Endogenous and Exogenous Factors Affecting Lipoprotein Lipase Activity

Home , APOA5, Contrast agent, Ibrolipim, Statin

Umeå University Medical Dissertations, New Series No. 1669

Endogenous and exogenous factors affecting lipoprotein lipase activity

Mikael Larsson

Department of Medical Biosciences, Physiological Chemistry Umeå 2014

Responsible publisher under Swedish law: the Dean of the Medical Faculty This work is protected by the Swedish Copyright Legislation (Act 1960:729) ISBN: 978-91-7601-115-7 ISSN: 0346-6612 E-version available at http:// http://umu.diva-portal.org/ Printed by: Cityprint I Norr AB Umeå, Sweden 2014 Abstract

Individuals with high levels of plasma triglycerides are at high risk to develop cardiovascular disease (CVD), currently one of the major causes of death worldwide. Recent epidemiological studies show that loss-of-function mutations in the APOC3 gene lower plasma triglyceride levels and reduce the incidence of coronary artery disease. The APOC3 gene encodes for apolipoprotein (APO) C3, known as an inhibitor of lipoprotein lipase (LPL) activity. Similarly, a common gain-of-function mutation in the LPL gene is associated with reduced risk for CVD.

LPL is central for the metabolism of lipids in blood. The enzyme acts at the endothelial surface of the capillary bed where it hydrolyzes triglycerides in circulating triglyceride-rich lipoproteins (TRLs) and thereby allows uptake of fatty acids in adjacent tissues. LPL activity has to be rapidly modulated to adapt to the metabolic demands of different tissues. The current view is that LPL is constitutively expressed and that the rapid modulation of the enzymatic activity occurs by some different controller proteins. Angiopoietin-like protein 4 (ANGPTL4) is one of the main candidates for control of LPL activity. ANGPTL4 causes irreversible inactivation through dissociation of the active LPL dimer to inactive monomers. Other proteins that have effects on LPL activity are the APOCs which are surface components of the substrate TRLs. APOC2 is a well- known LPL co-factor, whereas APOC1 and APOC3 independently inhibit LPL activity.

Given the important role of LPL for triglyceride homeostasis in blood, the aim of this thesis was to find small molecules that could increase LPL activity and serve as lead compounds in future drug discovery efforts. Another aim was to investigate the molecular mechanisms for how APOC1 and APOC3 inhibit LPL activity.

Using a small molecule screening library we have identified small molecules that can protect LPL from inactivation by ANGPTL4 during incubations in vitro. Following a structure-activity relationship study we have synthesized lead compounds that more efficiently protect LPL from inactivation by ANGPTL4 in vitro and also have dramatic triglyceride-lowering properties in vivo. In a separate study we show that low concentrations of fatty acids possess the ability to prevent inactivation of LPL by ANGPTL4 under in vitro conditions.

With regard to APOC1 and APOC3 we demonstrate that when bound to TRLs, these apolipoproteins prevent binding of LPL to the lipid/water interface. This results in decreased lipolysis and in an increased susceptibility of LPL to inactivation by ANGPTL4. We demonstrate that hydrophobic amino acid residues that are centrally located in the APOC3 molecule are critical for attachment of this protein to lipid emulsion particles and consequently for inhibition of LPL activity.

In summary, this work has identified a lead compound that protects LPL from inactivation by ANGPTL4 in vitro and lowers triglycerides in vivo. In addition, we propose a molecular mechanism for inhibition of LPL activity by APOC1 and APOC3.

i Table of Contents

Abbreviations iii

List of papers iv

Introduction 1

Transport and metabolism of exogenous lipids 2 Transport and metabolism of endogenous lipids 3 Reverse cholesterol transport 4

Dyslipidemia and cardiovascular disease 5

Determinants of plasma triglyceride metabolism 6

Lipoprotein lipase 6 Regulation of lipoprotein lipase activity 10 Lipoprotein lipase in atherosclerosis 11

Angiopoietin-like proteins 13

GPI-anchored HDL-binding protein 1 16

Apolipoproteins 17 Apolipoprotein A1 18 Apolipoprotein B 18 Apolipoprotein C1 19 Apolipoprotein C2 20 Apolipoprotein C3 22 Apolipoprotein E 24 Apolipoprotein A5 26

Aims of the thesis 27

Results and discussion 28 Paper I 28 Paper II 32 Paper III and IV 33

Conclusions 39

References 40

Acknowledgements 55

ii Abbreviations ABCA1 – ATP-binding cassette transporter A1 ANGPTL – angiopoietin-like protein APO – apolipoprotein ATP – adenosine triphosphate CETP – cholesteryl ester transfer protein CHD – coronary heart disease CVD – cardiovascular disease ER – endoplasmic reticulum GPI – glycosylphosphatidylinositol GPIHBP1 – GPI-anchored HDL-binding protein 1 HDL – high-density lipoprotein HTS – high-throughput screening IDL – intermediate-density lipoprotein LCAT – lecithin-cholesterol acyltransferase LDL – low-density lipoprotein LDLR – low-density lipoprotein receptor LMF1 – lipase maturation factor 1 LPL – lipoprotein lipase LRP1 – low-density lipoprotein receptor-related protein 1 LXR – liver X receptor LY6 – lymphocyte antigen 6 MTP – microsomal triglyceride transfer protein PPARs – peroxisome proliferator-activated receptors PTLP – phospholipid transfer protein SAR – structure–activity relationship SR-B1 – scavenger receptor class B member 1 TG – triglyceride TRL – triglyceride-rich lipoproteins VLDL – very low-density lipoprotein

iii Paper I

Apolipoproteins C-I and C-III inhibit lipoprotein lipase activity by displacement of the enzyme from lipid droplets

Larsson, M., Vorrsjö, E., Talmud, P., Lookene, A., and Olivecrona, G. (2013) The Journal of Biological Chemistry 288(47):33997-4008

Paper II

Fatty acids bind tightly to the N-terminal domain of angiopoietin-like protein 4 and modulate its interaction with lipoprotein lipase

Robal, T., Larsson, M., Martin, M., Olivecrona, G., and Lookene, A. (2012) The Journal of Biological Chemistry 287(35):29739-52

Paper III

Identification of a small molecule that stabilizes lipoprotein lipase in vitro and lowers triglycerides in vivo

Larsson, M., Caraballo, R., Ericsson, M., Lookene, A., Enquist, P. A., Elofsson, M., Nilsson, S. K., and Olivecrona, G. (2014) Biochemical and Biophysical Research Communications 25;450(2):1063-9

Paper IV

Structure-activity relationships of small molecules lowering plasma triglycerides

Caraballo, R., Larsson, M., Nilsson, S.K., Ericsson, M., Qian, W., Nguyen, N.P., Kindahl, T., Svensson, R., Mastej, M., Artursson, P., Olivecrona, G., Enquist, P.A., and Elofsson, M. Manuscript

iv Introduction

Triglycerides (or triacylglycerols) are the most abundant dietary lipids. They are used as source of energy in most tissues or for storage in adipose tissue. Triglycerides are non-polar esters made up of glycerol and long-chain fatty acids that are incapable of entering cells on their own. Hydrolysis of the ester bonds, catalysed by enzymes called lipases is therefore needed. By the action of lipases fatty acids are released from the glycerol backbone so that the polar lipolysis products (fatty acids and monoglycerides) can cross the plasma membrane of cells and be used for metabolic purposes. Besides serving as a source of energy, fatty acids are active substances and function as signal molecules. When in excess, fatty acids may lead to cellular dysfunction and even cell death. In contrast triglycerides are inert. Therefore fatty acids in excess are re-esterified to form triglycerides that in turn enable safe storage in intracellular lipid droplets and/or transport in lipoproteins in blood.

Lipoproteins are macromolecular assemblies of lipids and proteins composed of phospholipids and cholesterol that form spherical monolayers covering a core of triglycerides and cholesteryl esters. The polar headgroups of the phospholipids and the hydroxyl groups of cholesterol compose a hydration shell that surrounds the hydrophobic core. Apolipoproteins (APOs) are specific protein components of the surface layer of lipoproteins. They regulate lipoprotein metabolism by serving as receptor ligands and cofactors/inhibitors of enzymes. Lipoproteins are divided into subclasses based on their density (Table 1). The different lipoprotein classes have distinct origins and functions and compose a dynamic system that maintains lipid homeostasis in blood. Chylomicrons and VLDL are the largest lipoproteins. They are responsible for the transport of triglycerides in blood and are therefore important carriers of energy to cells, while LDL and HDL mainly serve as carriers of cholesterol.

Disturbances in lipid homeostasis are associated with common human diseases such as obesity, insulin resistance and diabetes. Ultimately, lipid disorders may lead to cardiovascular disease with fatal outcomes.

1 Chylomicrons VLDL IDL LDL HDL

0.95- 1.006- 1.019- 1.063- Density (g/cm3) <0.95 1.006 1.019 1.063 1.21 Diameter (nm) 75-1200 30-80 25-35 18-25 5-12 Chemical composition (% dry weight) Protein 1-2 10 18 25 33

Triglyceride 83 50 31 10 8 Cholesterol and 8 22 29 46 30 cholesteryl ester Phospholipid 7 18 22 22 29

Apolipoproteins B48, A, C, E B100, C, E B100, C, E B100 A, C, E

Table 1. Characteristics and composition of human lipoprotein classes [1].

Transport and metabolism of exogenous lipids After a meal, dietary triglycerides enter the gut in large insoluble lipid droplets. Bile salts help to emulsify these lipid droplets into smaller entities and thereby increase the accessible surface. The protein colipase binds and promotes hydrolysis of the triglycerides by pancreatic lipase [2]. The lipolysis products (fatty acids and monoglycerides) are taken up by intestinal absorptive cells (enterocytes) and subsequently resynthesized to triglycerides.

With the aid of microsomal transfer protein (MTP) and APOB48 triglycerides are incorporated into chylomicrons [3]. Chylomicrons are the largest of the lipoproteins and transport dietary triglycerides, fat-soluble vitamins, cholesterol and cholesteryl esters (Table 1). Compared to other lipoproteins, chylomicrons are large in size because of their massive triglyceride core. Due to their size chylomicrons cannot pass into the fenestrated capillaries of the intestinal mucosa. Consequently, chylomicrons are directed to the lymphatic system before entering the bloodstream via the left subclavian vein [4]. In blood, chylomicrons acquire additional apolipoproteins (APOCs and APOE) from high-density lipoproteins (HDL), which function as an apolipoprotein reservoir [5,6]. In capillaries chylomicrons bind avidly to a membrane- bound protein complex composed of lipoprotein lipase (LPL) and GPI-anchored HDL-binding protein 1 (GPIHBP1) [7]. LPL in turn is activated by APOC2 on the chylomicron surface and triglycerides from the core are readily hydrolyzed allowing delivery of lipolysis products to the underlying parenchymal cells [6,7]. During intravascular lipolysis the chylomicron particle decreases in size and is transformed into a chylomicron remnant. The chylomicron remnant particle is remodeled by release of excess surface material to HDL while mainly APOB48 and APOE remain bound to the remnant surface [8,9]. The small size of the chylomicron remnants allow them to pass the fenestrated capillary endothelium in the liver followed by cellular uptake via binding of APOE to members of the LDL (low-density lipoprotein) receptor

2 family [9]. A brief summary and schematic representation of chylomicron metabolism is shown in Figure 1.

Figure 1. Chylomicron metabolism.

Transport and metabolism of endogenous lipids The chylomicron is taken up in the liver by receptor-mediated endocytosis after the majority of its core triglycerides have been depleted by intravascular lipolysis. Constituents from the chylomicron remnant such as remaining triglycerides and dietary cholesterol, together with lipids endogenously synthesized in the liver, are utilized in the hepatocytes for formation of very low-density lipoprotein (VLDL). VLDL is formed by the stepwise lipidation of APOB100 with the aid of MTP [10]. After secretion from the liver the VLDL particle is rich in triglycerides and is hydrolyzed by LPL in capillaries [7]. During lipolysis, VLDL is transformed to an intermediate-density lipoprotein (IDL). Consequently surface rearrangements must occur and APOCs will detach from the surface while APOE is less affected and remain on the

IDL particle [5,11]. IDL can be removed from the circulation by APOE/APOB100-mediated uptake in the liver [12]. Alternatively, IDL can be furthered depleted of its triglyceride content by the enzyme hepatic lipase [13]. When core triglycerides in IDL are depleted, APOE will detach and the remaining particle is referred to as low-density lipoprotein (LDL) [14]. In contrast to the parental VLDL particle, LDL has a high cholesteryl ester to triglyceride ratio. LDL delivers cholesterol to extrahepatic tissues via endocytosis mediated via LDL receptors

3 [15], followed by hydrolysis of all constituents in the lysosomes. Uptake of LDL also occurs in the liver where the cholesterol content is incorporated in VLDL or excreted to the gut via bile [16,17]. A brief summary and schematic representation of the endogenous lipoprotein metabolism is shown in Figure 2.

Figure 2. Endogenous lipoprotein metabolism.

Reverse cholesterol transport HDL functions as a reservoir for APOCs and APOE in both the exogenous and the endogenous pathways for lipoprotein metabolism. Another important role for HDL is to accept excess cholesterol from extrahepatic tissues for transport to the liver where the cholesterol can be excreted in bile. This HDL-mediated process is called reverse cholesterol transport. Nascent HDL is formed in enterocytes and hepatocytes by the addition of phospholipids and free cholesterol to APOA1 via the ATP-binding cassette transporter A1 (ABCA1) [3,18]. In the circulation, nascent HDL is remodeled by phospholipid transfer protein (PTLP) which transfers phospholipids to the maturing HDL particle from excess surface coats of chylomicron remnants [19]. The phospholipid-rich nascent HDL becomes fully matured by the acquisition of free cholesterol from extrahepatic tissues by ATP-binding cassette transporters [20]. The majority of the mobilized cholesterol is esterified to cholesteryl esters by the action of HDL-bound lecithin-cholesterol acyltransferase (LCAT) and thereby relocalized from the surface of the HDL particle to the core [21]. Mature HDL release free

4 cholesterol and cholesteryl esters in the liver via binding to scavenger receptor class B1 (SR- B1) [22]. Alternatively, HDL can be relieved of cholesteryl esters by cholesteryl ester transfer protein (CETP) which exchange HDL-cholesteryl esters for triglycerides from other lipoprotein classes [23].

Dyslipidemia and cardiovascular disease

Cardiovascular disease (CVD) includes conditions that narrows and block blood vessels, often as a consequence of atherosclerosis and/or high blood pressure, leading to life-threatening events such as heart attacks or strokes [24]. Inflammation and impaired lipid metabolism (dyslipidemia) severely increases the risk of developing atherosclerosis. The onset of atherosclerosis is believed to be due to discrete areas of chronic inflammation in large and medium sized arteries due to infiltration and retention of LDL particles inside the blood vessel wall. A cascade of events follows that ultimately lead to the recruitment of monocytes that differentiates into macrophages which internalize lipoproteins and are transformed into lipid-loaded foam cells – the archetypical cell in atherosclerosis [reviewed [25,26]].

Patients at increased cardiovascular risk commonly display high levels of plasma triglycerides, elevated LDL cholesterol, small dense LDL particles and low levels of HDL- cholesterol. Statin therapy effectively reduces CVD events in patients with elevated LDL- cholesterol. However, many patients remain at high cardiovascular risk even after optimal reductions in LDL-cholesterol [27]. Numerous case-control studies have established positive correlations between plasma triglyceride levels and CVD, even after adjustment for LDL- cholesterol and HDL-cholesterol [reviewed [28]]. However, whether triglycerides have a causal role in the development of CVD, or serve as a predictive biomarker, has been debated for decades. Similarly, a discussion is ongoing on the benefit of HDL-cholesterol which since long has been shown to correlate inversely with CVD risk [reviewed [29,30]].

The level of triglycerides in blood is the result of environmental factors in combination with common and rare variants of multiple genes that govern lipoprotein metabolism. Some genes known to be involved in lipoprotein catabolism have a large impact on the levels of plasma triglycerides, such as LPL or genes that have been proven essential for proper LPL functionality. Individuals that are homozygous or compound heterozygous for loss-of- function mutations in these genes are incapable of hydrolyzing triglyceride-rich lipoproteins (TRLs), i.e. chylomicrons and VLDLs. Consequently these patients display severe hypertriglyceridemia, often accompanied by low levels of HDL-cholesterol due to impaired intravascular lipolysis. Interestingly, from a disease perspective, these patients may develop pancreatitis due to chylomicronemia but they do not experience increased CVD risk. Instead evidence point toward an increased CVD risk for individuals with moderate hypertriglyceridemia [31,32]. The etiology of non-severe hypertriglyceridemia is complex and may involve heterozygous mutations in genes affecting LPL functionality in combination with other determinants that each may have small effects on plasma triglyceride levels [reviewed [33]]. The current belief is that elevated fasting plasma triglyceride levels are a

5 consequence of an accumulation of deleterious alleles in genes regulating lipoprotein catabolism. Conversely, reduced plasma triglyceride levels are believed to be due to an accumulation of loss-of-function mutations in genes regulating lipoprotein assembly or secretion, or genes responsible for inhibition of lipoprotein catabolism like APOC3. Thus, plasma triglyceride levels and CVD risks are presumed to be due to the combined effects of such deleterious and protective alleles [31].

Even though much is known about determinants for plasma triglyceride and HDL-cholesterol levels, there are still gaps in our knowledge. This is well exemplified by the recent outcomes from randomized control trials regarding CETP-inhibition and triglyceride lowering therapy. CETP-inhibition resulting solely in increased plasma HDL-cholesterol was shown to be ineffective in reducing recurrent cardiovascular events [34]. Similarly, administration of niacin as an add-on therapy to statins failed to reduce the incidence of CVD despite reductions in plasma triglyceride and LDL-cholesterol levels in combination with increased levels of HDL-cholesterol [35]. It should be mentioned that there are ongoing trials with other CETP-inhibitors, and that randomized control trials with fibrates as lipid lowering therapy have shown beneficial effects in the prevention of coronary events [29,36].

Thus, assessing CVD risk solely based on plasma triglyceride and HDL levels may have its caveats. However, numerous epidemiological studies infer an associated CVD risk for both elevated triglycerides and low levels of HDL-cholesterol. Presumably future discoveries will enable us to distinguish between benign and at-risk individuals with similar lipid profiles.

Determinants of plasma triglyceride metabolism

The catabolism of circulating TRLs is important in maintaining plasma triglyceride homeostasis. LPL is central for this function and acts as a gatekeeper by directing fatty acids from the circulation to underlying tissues. The demands of fatty acids differ between tissues and depend on nutrient status as well as the relative energy consumption. As mentioned in the beginning, excessive fatty acid delivery may lead to cellular dysfunction and death because fatty acids at high concentrations are toxic to cells. Thus, LPL must be carefully regulated to meet up with the tissue-specific metabolic demand. Below follows a summary of the currently known main determinants of plasma triglyceride metabolism.

Lipoprotein lipase

The effect of LPL was first described by Hahn who noticed that administration of heparin to dogs abolished the turbidity in blood associated with postprandial lipemia [37]. Later, Korn concluded that the “clearing factor” was a heparin-releasable lipase that hydrolyzed core triglycerides of large lipoproteins and consequently reduced the light-scattering in lipemic plasma, and hence the enzyme was later named lipoprotein lipase [38].

Homozygous loss-of-function mutations in the gene for LPL are rare and result in creamy blood plasma due to accumulated TRLs. Individuals that are heterozygous for these

6 mutations display diverse phenotypes ranging from normal to high levels of plasma triglycerides, suggesting an increased predisposition to hypertriglyceridemia when confounding factors are present [39]. There are numerous identified human mutations in the LPL gene [reviewed [40]]. Loss-of-function mutations include splice site mutations, frameshift mutations, and missense mutations affecting secretion, endothelial transport or the catalytic function of LPL [7]. Also, other mutations may be accompanied by increased plasma triglyceride levels and reduced levels of HDL-cholesterol [40]. In contrast, a common gain-of-function mutation, LPL-S447X, has been associated with increased pre- and post- heparin LPL activity, reduced levels of plasma triglycerides and increased levels of HDL- cholesterol [41,42]. Numerous epidemiological studies have suggested that S447X-carriers have a lower risk for CVD compared to non-carriers, while other investigations have not been able to confirm these observations [42].

LPL is a 448 amino acid residue glycoprotein that is mainly synthesized in heart, skeletal muscle and adipose tissue. Other tissues also produce LPL. Of special interest is LPL in macrophages which may promote atherosclerosis (discussed below). LPL catalyzes the ester hydrolysis of triglycerides in TRLs, resulting in the release of fatty acids and monoglycerides. LPL is a serine hydrolase and a member of the lipase gene family with highest sequence homology to endothelial lipase followed by hepatic lipase and pancreatic lipase [43]. Due to the lack of available 3-D structure for LPL or any of its closest relatives an in silico modelling was performed based on the crystal structure of pancreatic lipase [44]. By analogy with pancreatic lipase it is assumed that LPL is composed of two domains; a large N-terminal domain (residues 1-312) which contains the active site and a smaller C-terminal domain (residues 313-448) that is required for binding to lipid substrates [45]. Unlike pancreatic lipase, LPL is catalytically active only as a non-covalent homodimer [46]. It was suggested that the two LPL monomers are oriented in a head-to-tail fashion where the C-terminal domains bind to the surface of lipid particles and allow the opposing catalytic N-terminal domains to act on individual lipid molecules in the surface layer of lipoproteins (Figure 3) [45,47]. Evidence for this is that co-expression of two inactive forms of LPL, with loss-of- function mutations in the C-terminal and N-terminal domains, respectively, resulted in catalytically active LPL [48]. LPL has high affinity for heparin. This was early indicated by the release of lipolytic activity to blood upon intravenous heparin injections [37,38], and this property was used for the initial large-scale purifications of LPL from bovine milk [49]. Later it has been demonstrated that surface exposed clusters of positive charges located in both the N- and C-terminal domains of LPL interact with negatively charged sulphate groups in heparin [44]. By using heparin-Sepharose chromatography it was shown that LPL protein elutes at two distinct salt concentrations, approximately at 0.5 M and 1 M of NaCl respectively, and that catalytic activity was only associated with the 1 M NaCl fraction [50]. Analytical ultracentrifugation and sedimentation studies in sucrose gradients have shown that native LPL has a molecular mass of about 110 kDa, corresponding to a non-covalent homodimer of 55 kDa subunits [46].

Figure 3. Homology model of LPL based on the crystal structure of pancreatic lipase using SWISS-MODEL [51,52,53]. The N-terminal (orange) of one LPL monomer is presumed to interact with the opposing C-terminal (gray) of another LPL monomer, forming a head-to-tail homodimer.

LPL can hydrolyze triglycerides and phospholipids in all lipoprotein classes. However, when different substrates are presented in a mixture, such as in the plasma, triglycerides of TRLs are almost exclusively hydrolyzed, presumably due to the preferred binding of LPL to TRLs and to that triglycerides are a preferred substrate over phospholipids [54]. Cleavage of LPL by chymotrypsin at residues 390-391 and 392-393 abolished binding and enzymatic activity towards TRLs, while the activity towards the more water-soluble substrates tributyrin and para-nitrophenyl butyrate remained unaffected. These findings suggested a anchoring function of the C-terminal domain of LPL to the surface of the lipid droplets [55]. The chymotrypsin-truncated form of LPL was capable of binding to synthetic lipid emulsions although with severely impaired lipolytic capability and the authors proposed that the C- terminal domain is also crucial for a correct interfacial orientation of the enzyme. APOC2

8 increases the LPL activity on synthetic lipid emulsions composed of long-chain triglycerides, but has little or no effect with short-chain triglycerides (e.g. tributyrin) [56]. By the use of chimeras (hepatic lipase and lipoprotein lipase) it was suggested that APOC2 interacts directly both with the LPL C-terminal domain after residue 388 and with the N-terminal domain, presumably each on different subunits [57]. Other studies showed, however, that chymotrypsin-cleaved LPL was dependent on APOC2 and had a maximal activity of about 25% compared to that of intact LPL for lipid emulsions of long-chain triglycerides. Soluble ester substrates like tributyrin were hydrolyzed just as well with the chymotrypsin-cleaved LPL as they were with the intact enzyme [55].

When LPL hydrolyzes long-chain triglyceride substrates, the activity becomes markedly reduced with time unless a fatty acid acceptor is present (e.g. albumin) due to product inhibition. In the absence of fatty acid acceptors, the LPL reaction may seemingly come to a stop, but on addition of albumin the inhibition is immediately relieved [58]. The authors suggested that LPL forms a complex with fatty acids that prevents further hydrolysis of triglycerides, and that this could serve as a feed-back mechanism to prevent excessive fatty acid delivery to cells and tissues in vivo.

LPL contains two conserved N-linked glycosylation sites. In human LPL these are located at Asn 43 and Asn 359. Expression studies with mutated LPL in COS cells demonstrated intracellular accumulation within the endoplasmic reticulum (ER) and complete loss of enzymatic activity when Asn 43 was substituted for Ala. In contrast, the LPL mutant N359A was secreted, and showed only a minor loss in enzymatic activity compared to the wild-type protein [59,60]. The ability of cells to properly secrete active LPL depends on the unique ER membrane protein called lipase maturation factor 1 (LMF1). Recently it was shown that homozygous loss-of-function mutations in the LMF1 gene cause severe hypertriglyceridemia due extremely low LPL activity, combined with moderate reductions in hepatic lipase activity in post-heparin plasma [61]. It was suggested that LMF1 functions as a chaperone for the assembly of functional homodimers for LPL, endothelial lipase and hepatic lipase [62]. Unlike humans with LPL deficiency, Lpl-knockout mice die soon after birth. They can be rescued by transient expression of LPL from adenovirus during the neonatal stage [63,64]. Lpl-knockout pups are mildly hypertriglyceridemic immediately after birth but on suckling they manifest severe hypertriglyceridemia with markedly reduced levels of LDL- and HDL-cholesterol compared to wild-type pups. Heterozygous Lpl-knockout pups display only mild hypertriglyceridemia during suckling [63]. In contrast, transgenic mice that overexpress LPL display hypotriglyceridemia with increased LDL-cholesterol levels and a concomitant decrease in VLDL-cholesterol. These observations were suggested to reflect a more rapid conversion of VLDL to LDL [65].

Chylomicrons and large VLDLs do not readily move across the capillary endothelium. Thus, lipid delivery to cells depends on the activity of LPL that generates more polar lipolysis products, fatty acids and monoglycerides, which can cross cellular membranes. LPL also

9 promotes cholesterol delivery to cells by initiating the formation of LDL particles from parental VLDLs [66]. Released lipolysis products are predominantly taken up by parenchymal cells and used as energy source or stored as triglycerides in lipid droplets for later use. Tissues with a high energy demand, such as striated muscle tissues (heart and skeletal muscle), primarily use fatty acids as an energy source. Adipose tissue instead stores lipolysis products for later mobilization when dietary lipids are no longer available. To accommodate for efficient and appropriate fatty acid delivery, the activity of LPL is rapidly modulated in different tissues to readily adapt to their metabolic needs [54]. The rapid regulation of LPL activity is believed to mainly occur at the post-transcriptional level. The current knowledge regarding the tissue-specific regulation of LPL suggests two conceptually different mechanisms. In adipose tissue it appears that regulation occurs by irreversible inactivation of the enzyme, while in striated muscle the amount of active LPL that is located on the vascular side of the endothelium in contact with blood seems to be regulated. A summary of the tissue-specific regulation of LPL in rats follow below.

Regulation of lipoprotein lipase activity

In adipose tissue the nutrient status is presumed to be the main determinant of the level of LPL activity. In the fed state LPL activity is high to promote lipid storage. In contrast, the enzymatic activity is markedly reduced in the fasted state. By comparing fed and fasted rats, Olivecrona and coworkers showed that fasting had modest effects on LPL mass while the ratio of active LPL dimers over inactive monomers was substantially decreased with a concomitant loss of LPL activity in adipose tissue of the fasted animals [67]. Subsequent studies showed that administration of the transcription blocker actinomycin D to fasted rats rapidly reverted the fasting-induced reduction in LPL activity and this was accompanied by an increased ratio of active LPL dimers over inactive monomers [68]. These observations could not be explained by changes in LPL mRNA or mass. It was suggested that another protein, with a more rapid turnover, causes inactivation of LPL in adipose tissue on fasting. Later studies showed that angiopoietin-like protein 4 (ANGPTL4) could inactivate LPL by converting active dimers to inactive monomers. In addition, the expression of ANGPTL4 in adipose tissue correlated with LPL activity during both fed-to-fasted and fasted-to-fed transitions [69]. It now seems likely that ANGPTL4 is a key determinant for regulation of LPL activity in adipose tissue [70].

In contrast to adipose tissue, LPL activity in homogenates of rat hearts does not change much on fasting. More detailed studies have shown that the amount of heparin-releasable LPL activity increases during fasting [71,72]. An advantage with the heart is that it allows for extracorporeal perfusion experiments in which LPL bound to the endothelium facing the lumen can be released by a heparin-containing medium and then quantified [73]. Due to technical difficulties to perfuse adipose tissue, an alternative was to compare the LPL activity in total tissue to that of isolated adipocytes from the same tissue. With this method it was concluded that the specific LPL activity (activity/mass) was reduced only in the extracellular

10 fraction of rat adipose tissue upon fasting [74]. It is noteworthy that the extracellular fraction of LPL in adipose tissue represented approximately 70% of the total LPL pool. In following studies of the perfused rat heart it was shown that only 10-20% of the total LPL pool was released upon heparin addition [75]. It remains to be determined if the residual pool of LPL resides within the interstitial spaces or inside cardiomyocytes. The specific activity of LPL in homogenates of heart is not affected by nutrient status. Instead there was a shift in the amount of heparin-releasable LPL that was increased during fasting. Conversely, heparin-releasable LPL mass was low in the fed state but could be reverted by the administration of actinomycin D [75]. These observations suggest a regulatory mechanism involving a protein responsible for transporting LPL to its endothelial site of action, rather than an effect on the catalytic function of LPL as seen in adipose tissue. A strong candidate for this regulation is GPIHBP1 that has been found to transport LPL across the capillary endothelium [76]. More studies are needed to understand how GPIHBP1 transports LPL to the capillary lumen and if additional factors are involved in this process. The mechanisms for regulation of LPL activity in skeletal muscle is less understood than in heart and adipose tissue. Changes in LPL activity were reported to correlate with LPL mRNA and enzymatic activity has been found to increase during fasting. Several studies correlate acute exercise with increased expression of LPL and with increased LPL activity in skeletal muscle [71]. Hindlimb unloading (inactivity) demonstrated a dramatic reduction in LPL activity in the soleus muscle of rats compared with that in the muscles of ambulatory control rats. The observed loss in LPL activity could not be explained by a decrease in tissue LPL mass. However, the amount of heparin-releasable LPL mass was markedly reduced. The reduction in LPL activity associated with physical inactivity was prevented by administration of actinomycin D, but also reverted by treadmill walking [77]. Taken together, LPL regulation in skeletal muscle show several similarities to that of the heart and point toward GPIHBP1 as part of the regulatory process in times when there is a local energy demand.

Lipoprotein lipase in atherosclerosis

The role of LPL in atherosclerosis is dual-faceted. It was early postulated that the action of LPL may be pro-atherogenic because remnant lipoproteins, depleted of their triglyceride core, should be capable of entering the arterial wall while unprocessed large TRLs remain in the circulation [78,79]. In support of this presumption, individuals with homozygous loss-of- function mutations in the LPL gene, or in genes responsible for proper enzymatic functionality, are not associated with severely increased CVD risk [31]. However, the involvement of LPL in atherosclerosis is far more complex than originally anticipated. Individuals with heterozygous loss-of-function mutations in the LPL gene have been reported to be predisposed to premature atherosclerosis [80]. Although unprocessed circulating TRLs are not able to cross the endothelial barrier, evidence point toward involvement of remnant lipoproteins with not fully depleted triglyceride cores in the development of atherosclerosis. To further add complexity, it was shown that severely hypertriglyceridemic mice, deficient of

11 either LPL or GPIHBP1, spontaneously develop atherosclerosis on a normal diet [81,82]. Whether these observations are species-specific effects warrant further studies

It was suggested that lipolysis products derived from LPL may promote atherosclerosis by causing vascular injury and increased endothelial permeability. This was based on in vitro studies using aortic endothelial cells [83]. Previous studies had shown that injection of labeled human VLDLs followed by injections of LPL increased the amount of VLDL-label in the arterial wall of perfused rat carotid arteries compared to when no LPL was injected. Injections of HDL prior to injections of the labeled VLDLs and LPL prevented the VLDL-label to enter the arterial wall [84]. Assuming that LPL is not present at the luminal endothelium of arteries, due to the lack of GPIHBP1 in these vessels [7], the pro-atherogenic effects of lipolysis products should be due to either effects of lipolysis spillover in adjacent tissues [reviewed [85]] – or to that catalytically active LPL within the arterial wall (e.g. macrophage LPL) may account for such effects by hydrolysis of triglycerides in retained remnant lipoproteins. Irrespectively, the observation regarding the protective effects of HDL is highly interesting. Possibly a similar mechanism could account for the anti-atherogenic effects seen by overexpression of LPL in atherosclerotic mice models that lacks the LDL-receptor (LDLR) or its ligand APOE. When fed an atherogenic diet, the combined Ldlr-knockout and LPL- transgenic mice displayed an 18-fold reduction in atherosclerotic mean lesion area compared to Ldlr-knockout mice. As expected, overexpression of LPL increased HDL- cholesterol levels and erased the relatively mild hypertriglyceridemia associated with the LDLR deficiency. Interestingly, LPL overexpression reduced plasma LDL-cholesterol levels by approximately 50% without effects on circulating APOB levels [86]. Similar experiments on Apoe-knockout mice showed a more modest two-fold reduction in mean lesion area upon LPL overexpression. The lesion area size was in the same order of magnitude as that observed for Ldlr-knockout mice in the previous study. Analogous to the previous study the combined Apoe-knockout and LPL-transgenic mice showed lower plasma triglycerides levels compared to Apoe-knockout mice while there were no differences in circulating APOB levels. However, in contrast to the Ldlr-knockout mice, total cholesterol levels remained unaffected [87]. The detailed mechanism behind why LPL overexpression protects against atherosclerosis is intriguing but remains unknown. However, it is striking that cholesterol levels were only affected in Ldlr-knockout mice who also benefitted dramatically by LPL overexpression in terms of reduced atherosclerotic lesion area. Studies on transgenic rabbits overexpressing LPL show conflicting results. One study showed that overexpression of LPL protected against diet-induced atherosclerosis [88]. The same group later showed that overexpression of LPL promoted atherosclerosis [89]. Their suggestion for these discrepancies were that in the first study reduced lesion formation was a result of reduced plasma cholesterol in the transgenic animals, while in the second study cholesterol levels were similar between transgenic and wild-type rabbits. The authors proposed that the anti- atherogenic effects of LPL are due to increased remnant removal while the pro-atherogenic effects of LPL are caused by an increased amount of atherogenic LDL particles due to a more rapid and extensive depletion of VLDL-triglycerides. The latter suggestion is in contrast to the

12 observations seen in mice deficient of either LDLR or APOE, as discussed above. Possibly there are confounding effects which differ among species.

In contrast to the anti-atherogenic properties of endothelial bound LPL in mice, LPL in macrophages is presumed to be pro-atherogenic. By ablating endogenous macrophages in Ldlr-knockout mice using lethal irradiation, Babaev et al. showed that transplantation of macrophages from an LPL-deficient donor reduced the mean lesion area by approximately 30% when compared with irradiated littermates transplanted with macrophages which had normal LPL expression levels [90]. In this study the lipid profiles were not different between the groups of mice, suggesting that macrophage transplantation had no impact on circulating TRLs. Reductions in mean lesion areas were seen upon irradiation followed by macrophage transplantations in wild-type C57BL/6 mice [91,92]. It should be mentioned that these mice only developed minor lesions even when LPL-expressing macrophages were transplanted. Gustafsson et al. were successful in reproducing the observations previously made by Babaev et al. with macrophage transplantation from LPL-deficient donors into irradiated Ldlr-knockout mice [93]. They also showed that LDL retention in the aortae was reduced in mice with LPL-deficient macrophage transplants compared to those transplanted with LPL-expressing macrophages. They overexpressed both catalytically active and inactive LPL in macrophages on an APOE-deficient background. Modest overexpression of macrophage LPL in Apoe-knockout mice increased lesion size area by approximately 30% independently of catalytic ability, suggesting that the main effects are due to the non- catalytic properties of LPL [93]. Wu et al. showed that local transient overexpression of either catalytically active or inactive LPL in the carotid artery of rabbits receiving a normal diet resulted in rapid lipid deposition in the arterial wall [94]. It is evident that LPL exerts pro- atherogenic functions in the arterial wall, mainly by non-enzymatic effects. LPL can mediate binding and subsequent uptake of atherogenic lipoproteins in macrophages independent of LDLR [95]. It was also shown that LPL can bridge lipoproteins to the extracellular matrix, suggesting that LPL could promote retention of atherogenic lipoproteins in the artery wall [96].

Angiopoietin-like proteins

Several lines of evidence implicate that members of the angiopoietin-like protein family (ANGPTL3, -4, -8) are important LPL modulators. Transgenic mouse models overexpressing either ANGPTL3, -4 or -8 all display increased circulating triglyceride levels, while corresponding knockout models all show reduced triglyceride levels compared to wild-type mice [71]. Importantly, ANGPTLs differ in their expression, regulation and tissue distribution suggesting that they have distinct roles in lipid metabolism. In support of this is that Angptl3- knockout mice show increased LPL activity only in the fed state while the increase in LPL activity for Angptl4-knockout mice is more pronounced in the fasted state [97]. This suggests that the ANGPTLs may act as control proteins for LPL activity but under different metabolic conditions.

13 ANGPTL4 is under the transcriptional control of peroxisome proliferator-activated receptors (PPARs) [98,99]. Fatty acids are PPAR agonists, suggesting a feed-back mechanism involving ANGPTL4 with the aim to control lipid homeostasis [100,101,102]. Mice display a complex regulation of ANGPTL4 partly due to overlapping functions between PPAR isoforms. ANGPTL4 is synthesized by numerous tissues in mice, including white and brown adipose tissue, skeletal muscle, heart, liver and macrophages, supporting that it has a role as a paracrine factor to control LPL activity [71]. Hepatocytes do not normally produce LPL, while ANGPTL4 is mainly produced by the liver in humans. ANGPTL4 is found in blood, and intravenous injection of recombinant ANGPTL4 to mice, or overexpression of ANGPTL4 in mouse livers, have effects on LPL activity in post-heparin plasma, suggesting effects on the total tissue pool of LPL [97,103]. Thus, an endocrine role of ANGPTL4 circulating in blood cannot be excluded [71]. There are conflicting results on whether ANGPTL4 may affect hepatic lipase activity. Transgenic overexpression of ANGPTL4 in mice showed no effects on post-heparin hepatic lipase activity in one study, while a 50% reduction was found in another study [97,104]. There are no reports whether ANGPTL4 have effects on endothelial lipase.

ANGPTL4 consists of two domains, an N-terminal domain containing a coil-coiled region, followed by a C-terminal fibrinogen-like domain. Both the full-length protein and the N- terminal domain can inhibit LPL activity, while the C-terminal domain alone has no inactivating effect on LPL [99,103]. The mechanism for LPL inhibition is presumed to be due to conversion of active LPL dimers to inactive monomers [69]. Recently, another study suggested that ANGPTL4 inhibits LPL via a non-competitive, reversible mechanism and thereby prevents LPL to reach its site of action in vivo [105]. It is noteworthy that in this study deoxycholic acid was used to prevent the spontaneous inactivation of LPL. In the presence of other stabilizing factors like substrate lipoproteins, LPL is less susceptible to become inactivated by ANGPTL4 [106].

Generation of Angptl4-knockout mice, ANGPTL4-transgenic mice as well as injection of recombinant ANGPTL4 or injection of an neutralizing antibody against ANGPTL4 to different mouse models infer effects on LPL in vivo [97,103,107]. Interestingly, intercrossing Angptl4- knockout mice with atherosclerotic-inducible Apoe-knockout mice suppressed foam cell formation and protected against atherosclerosis compared with APOE-deficient control mice [108]. In contrast, Angptl4-knockout mice given a diet high in saturated fat developed a lethal phenotype with lipid-laden (foamy) macrophages in the mesenteric lymph nodes [101]. The authors proposed that ANGPTL4 prevents macrophage activation by reducing fatty acid uptake from triglycerides in chyle mediated by LPL. Another study showed that ANGPTL4 mimics the effects of tetrahydrolipstatin, an active-site inhibitor of LPL [109], regarding reduced uptake of oxidized LDL in macrophages [110]. In the same study, transgenic mice overexpressing ANGPTL4 were intercrossed with atherosclerosis-prone apolipoprotein E*3-Leiden mice. Combined ANGPTL4-transgenic and E*3-Leiden mice showed no difference in atherosclerosis susceptibility but displayed a 34% reduction in total lesion area size compared to E*3-Leiden mice. Previous studies have demonstrated that LPL

14 activity is under control of ANGPTL4 in macrophages [111]. Together these studies raise the possibility that the catalytic activity of LPL in macrophages may have pro-atherogenic effects in terms of atherosclerotic lesion progression.

In population studies it was found that a missense mutation in the ANGPTL4 gene, E40K, was present in approximately 3% of European Americans and resulted in reduced levels of plasma triglycerides and increased levels of HDL-cholesterol [112]. Later studies showed that a recombinant N-terminal fragment of ANGPTL4 containing the E40K mutation was unable to bind to LPL explaining the lack of function [113]. Other rare mutations within the ANGPTL4 gene have also been associated with reduced levels of plasma triglycerides [114,115]. Although loss-of-function mutations in the ANGPTL4 gene are associated with an anti-atherogenic lipid profile, there are conflicting results regarding the beneficial effect of these mutations on cardiovascular outcomes. One study showed that E40K-ANGPTL4 carriers were protected from coronary heart disease (CHD) compared to non-carriers [116]. In contrast, another study showed an association between E40K-ANGPTL4 and increased risk for CHD [115]. Further studies are needed to validate the role of loss-of-function mutations in the gene of ANGPTL4 and their implications for CVD. In addition to effects on lipid metabolism, ANGPTL4 has other non-metabolic functions, presumably linked to the C- terminal fibrinogen-like domain [reviewed [117]].

ANGPTL3 was the first member of this gene family to be recognized for its involvement in lipid metabolism. A mutant strain of KK obese mice (KK/San) showed marked reductions in plasma free fatty acids, and levels of total cholesterol (including HDL-cholesterol) and triglyceride compared to KK mice. It was later found that the hypotriglyceridemic mice were homozygous for a mutation in the Angptl3 gene and that mutations in the human ortholog gene also gave rise to hypolipidemia [118,119]. Fasted Angptl3-knockout mice displayed approximately 50% increased post-heparin LPL and hepatic lipase activities compared to wild-type mice [120]. Another study showed opposite effects, where Angptl3-knockout mice had slightly increased LPL activity in the fasted state while activity was dramatically increased in the fed state compared to wild-type mice [97]. ANGPTL3 was also reported to inhibit endothelial lipase activity and suggested to account for the low levels of HDL- cholesterol seen in ANGPTL3-deficient mice [121].

The biochemical mechanism for how ANGPTL3 inhibits LPL activity is unclear. Compared to ANGPTL4, several-fold more ANGPTL3 was required to cause LPL inactivation [122]. Based on kinetic studies it was suggested that ANGPTL3 causes inhibition of LPL rather than irreversible inactivation of the enzyme [123]. It was proposed that ANGPTL3 increases the susceptibility of LPL to become cleaved by proprotein convertases [124]. Similar effects were reported for ANGPTL4 [125]. One possible explanation for this could be that LPL is more prone to cleavage by proteinases in its monomeric conformational state [126]. LPL monomers are formed on inactivation by ANGPTL4 and these data suggests that also ANGPTL3 may inactivate LPL by a dimer to monomer conversion.

15 ANGPTL3 is exclusively expressed in the liver both in mice and humans. Like ANGPTL4, ANGPTL3 is found in blood. This suggests the effects of ANGPTL3 on the LPL system is accomplished by the circulating protein. Due to the modest effects of ANGPTL3 on LPL in the presence of TRLs [106], it is hard to understand the mode of action of ANGPTL3 in vivo. It was recently suggested that ANGPTL3 acts in concert with ANGPTL8 [127]. ANGPTL8 is expressed in liver, brain and adipose tissue in mice and humans. Interestingly, the expression of ANGPTL8 is increased by feeding and decreased during fasting (in mice), suggesting a role of ANGPTL8 as a LPL controller in the fed state. More studies are needed to corroborate these observations, and to unravel the biochemical mechanisms for the effects of the ANGPTLs on the LPL system, both in mice and humans.

GPI-anchored HDL-binding protein 1

Previous dogma held that LPL interacted with heparane sulphate proteoglycans at its site of action on the luminal capillary endothelium. This view has been changed with the discovery of GPIHBP1, a GPI-anchored protein of capillary endothelial cells, and the evidence for the importance of GPIHBP1 for plasma triglyceride metabolism [76]. It was clearly demonstrated by Stephen Young and coworkers that GPIHBP1 binds to LPL in the subendothelial spaces and shuttles the enzyme to the capillary lumen where it can interact with TRLs. When GPIHBP1 is absent, LPL remains in the interstitial spaces. This mislocalization of LPL results in severe hypertriglyceridemia [128]. Recently it was shown that the complex of LPL and GPIHBP1 on endothelial cells is crucial for the so called margination of TRLs, referring to the attachment of the TRLs to the luminal side of the endothelium allowing triglyceride hydrolysis by LPL to proceed [129]. After the discovery of GPIHBP1, several human loss-of- function mutations in the gene of GPIHBP1 have been reported [reviewed [130]]. Common for these mutations are that homozygous carriers have extremely low post-heparin LPL activity in plasma, and that they display a similar phenotype as those with LPL-deficiency. Heterozygous carriers have approximately 50% of the LPL mass and corresponding activity in post-heparin plasma compared to non-carriers. The plasma triglyceride levels are however usually within the normal range [131]. This is an interesting observation suggesting that GPIHBP1 is not rate-limiting in terms of plasma triglyceride removal, but it does indicate that the subendothelial pool of LPL is in excess compared to GPIHBP1. These facts opens up for that the transport of LPL to the capillary lumen might very well serve as a regulatory element for controlling fatty acid delivery to cells in striated muscle, as discussed previously.

Human GPIHBP1 is made up of two domains followed by a GPI-anchor which attaches the mature protein to the endothelial cell surface layer. The N-terminal domain consists of a stretch of 25 amino acid residues highly enriched in negatively charged aspartic and glutamic acid residues. Deletion of this domain or alanine-replacements of all negatively charged amino acid residues in the N-terminal domain, abolished interaction with LPL. In the C- terminal domain 10 cysteine residues are highly conserved across species. These residues are presumed to form intramolecular disulphide bonds, similar to those observed in UPAR/ly6 proteins. By cysteine to alanine replacement studies it was found that each

16 individual cysteine residue was essential for the interaction with LPL but not for secretion of GPIHBP1 to the cell surface. In addition, it was found that an epitope adjacent to the N- terminal acidic domain was still exposed in loss-of-function mutations in the C-terminal ly6- domain [132,133]. These observations suggest that the C-terminal ly6-domain binds directly to LPL, alternatively that it positions the N-terminal acidic domain in such a way that it can interact with LPL.

Gpihbp1-knockout mice are severely hypertriglyceridemic due to the mislocalization of LPL within interstitial spaces [128]. When mice that were genetically manipulated to express LPL in endothelial cells were intercrossed with Gpihbp1-knockout mice, the margination of TRLs was not stimulated, while plasma triglycerides were markedly reduced. The authors suggested that this was due to lipolysis by LPL secreted to the blood stream, rather than to lipolysis ocurring on the plasma membranes of the endothelial cells. Interestingly, intercrossing Gpihbp1-knockout mice with Angptl4-knockout mice resulted in offspring with mild hypertriglyceridemia. It was shown that GPIHBP1 protects LPL against both ANGPTL3- and ANGPTL4-mediated inactivation under in vitro conditions. In addition, administration of an antibody against ANGPTL4 in fasted Gpihbp1-knockout mice almost normalized their triglyceride levels while the effects were much less pronounced using a corresponding antibody against ANGPTL3 in fed Gpihbp1-knockout mice [122]. What is peculiar about these findings is that in the absence of GPIHBP1 no LPL should be transported to the capillary lumen. Another mystery is that in contrast to humans, LPL can be slowly released to blood by heparin in Gpihbp1-knockout mice. The release of LPL is markedly delayed in these mice. Yet, LPL mass and specific activity is similar to wild-type mice after 10 minutes, suggesting that the LPL pool is not inactivated [134]. With this in mind it is hard to understand how ANGPTL4-deficiency could ameliorate the hypertriglyceridemia associated with GPIHBP1- deficiency. Another possibility for such an observation could be that ANGPTL4 enables LPL to passively travel across the endothelial barrier to the capillary lumen.

Apolipoproteins

Apolipoproteins constitute the protein content of lipoproteins (see Table 1) and are located on the surface of the lipoprotein particles. The APOBs initiate lipoprotein assembly and are present in one copy per lipoprotein particle and remain attached during the entire metabolic cycle of the lipoprotein. In contrast, the other major apolipoproteins (APOAs, APOCs and APOE) are not permanently attached to one lipoprotein particle. Instead, these apolipoproteins are frequently exchanged among the lipoproteins and function as receptor ligands and enzyme cofactors/inhibitors. The exchangeable apolipoproteins share a secondary structural motif referred to as an amphipathic α-helix [135]. The amphipathic α- helix has a non-polar face which is presumed to associate with the proximal parts of the fatty acyl chains of the surface phospholipids on the lipoproteins. On the opposing face of the helix there are charged amino acid residues disposed such that they can interact with solvent molecules and the polar headgroups of the phospholipids [136,137]. The

17 amphipathic properties of the α-helix prevent apolipoproteins to bury into the hydrophobic environment of lipoproteins. As a result, the apolipoproteins remain at the lipid/water interface which facilitates transfer between lipoproteins and enables interaction with enzymes and cell-surface receptors [138].

In humans the genes for the main exchangeable apolipoproteins are primarily clustered at two distinct chromosomal regions. It was suggested that apolipoproteins have diverged through evolution from a common ancestral protein [139]. The genes (APOA1/C3/A4/A5) and (APOE/C1/C4/C2) are located on chromosome 11 and 19 respectively. The gene for APOA2 is on chromosome 1 and the gene for the main non-exchangeable apolipoprotein APOB is on chromosome 2 [138]. Below follows a brief description of the most studied apolipoproteins with focus on the APOCs and their relationship to LPL. In addition, another apolipoprotein, APOA5, is discussed for its particular, and yet not fully understood, importance for plasma triglyceride metabolism.

Apolipoprotein A1

Individuals with a premature stop mutation in the gene of APOA1 have markedly decreased levels of HDL-cholesterol and develop premature CHD [140]. APOA1 is mainly synthesized in the liver where it interacts with ABCA1 which aids in the formation of nascent HDL by the addition of phospholipids and cholesterol. APOA1 is also synthesized by enterocytes and becomes associated with chylomicrons. During intravascular lipolysis HDL is matured by the release of chylomicron surface material along with APOA1 in a process that depends on both LPL and PTLP [141,142]. The role of LPL for HDL metabolism is complex [as discussed [143]]. HDL has major functions in reverse cholesterol transport as acceptors for cholesterol, mediated from cells by ABC-transporters. APOA1 activates LCAT which enables incorporation of cholesterol into the lipid core of HDL through conversion of free cholesterol to cholesteryl esters [144]. In addition, APOA1 acts as a ligand for SR-B1 and thus facilitates cholesterol uptake in the liver [145].

Apolipoprotein B

APOB is synthesized in both liver and intestine where it initiates intracellular lipoprotein assembly with the aid of MTP. Liver-specific expression of APOB (APOB100) translates into a polypeptide chain with 4536 amino acid residues, while the intestine expresses an truncated form corresponding to approximately 48% of the amino acid sequence of the full-length protein. Hence, it is referred to as APOB48 [146]. Consequently, in humans chylomicrons are assembled with APOB48 while VLDLs are assembled with APOB100 [146]. In contrast, mice synthesize APOB48 lipoproteins also in the liver [147]. The main difference between the two isoforms is that APOB48 lacks the ability to interact with members of the LDL-receptor family, and therefore cannot be internalized into cells via this pathway. In order for chylomicron remnants to interact with members of the LDL-receptor family these lipoproteins must carry

APOE. The importance of APOB100 is highlighted in patients suffering from a condition

18 referred to as familial defective apolipoprotein B100. These individuals display hypercholesterolemia and have increased risk for developing premature atherosclerosis due nonsense mutations in the gene of APOB resulting in a truncated protein of 3500 amino acid residues [146].

Apolipoprotein C1

There are no known monogenic loss-of-function mutations in the human APOC1 gene.

APOC1 is a 57 amino acid residue polypeptide primary synthesized in the liver. APOC1 is a constituent of chylomicrons, VLDL and HDL in the circulation [148]. APOC1 inhibits LPL activity in vitro, but the mechanism for this effect has not been previously described [149]. Structurally, APOC1 is composed of two long amphipathic α-helices both with polar faces composed of negative residues (Figure 4) [150]. Point mutations in either of the two helices strongly influence the ability of APOC1 to bind to, to modify and to be retained on lipoprotein-like structures [151]. There are currently no reported mutagenesis studies on APOC1 that examine effects on LPL inhibition. It was shown that APOC1 can interfere with

APOE-dependent, but not APOB100-dependent, uptake of remnant lipoproteins. These observations were suggested to be the result of either conformational changes of APOE or displacement of this protein from the lipoprotein by the presence of APOC1 [152,153].

Surprisingly, Apoc1-knockout mice display slightly elevated plasma triglyceride levels and are more prone to develop hypercholesterolemia compared to wild-type mice when challenged with a severe atherogenic diet, suggesting impaired lipoprotein remnant clearance [154]. Early studies on APOC1-transgenic mice showed increased plasma cholesterol and triglyceride levels compared to wild-type mice. The effect of APOC1 was presumed to be due to impaired clearance of VLDL [155]. APOC1-transgenic mice that were intercrossed with Apoe-knockout mice displayed increased plasma cholesterol and triglyceride levels compared to Apoe-knockout mice. However, triglyceride levels were several-fold more increased than the corresponding cholesterol levels. The authors suggested that the main effect of overexpression of APOC1 was due to inhibition of LPL [156]. Animal studies also suggest that physiological levels of APOC1 have effects on plasma lipid levels. By comparing combined Apoe/Apoc1-knockout mice with Apoe-knockout mice it was proposed that endogenous levels of APOC1 increase VLDL production and inhibit LPL activity [157]. APOC1 was also suggested to be involved in reverse cholesterol transport by acting as an activator for LCAT [144] and as an inhibitor of CETP in humans [158].

Figure 4. High-resolution structure of human APOC1 in complex with sodium dodecyl sulfate micelles [150].

Apolipoprotein C2 Homozygous loss-of-function mutations in the human gene of APOC2 are rare and manifests in chylomicronemia and reduced levels of HDL-cholesterol, almost as severe as seen on LPL deficiency [159,160]. Several mutations that result in low or absent plasma levels of APOC2 have been reported and include premature stop mutations, donor splice-site mutations, exon missense mutations or null mutations within the promotor region [reviewed [161]]. Individuals heterozygous for loss-of-function mutations in the APOC2 gene display normal plasma triglyceride levels, unless genetic or environmental confounding factors are present [162]. Interestingly, one case of drug-resistant hypertriglyceridemia was reported and suggested to be due to high levels of plasma APOC2 [163].

APOC2 is a 79 amino acid residue polypeptide primarily synthesized in the liver. APOC2 is present on chylomicrons, VLDL and HDL [6,164]. High concentration of APOC2 inhibits LPL activity in vitro, but the explanation for these observations has not been studied in detail [149]. At physiological concentrations, APOC2 increases the activity of LPL on TRLs and emulsified long-chain triglyceride substrates [56,165]. However, there are important differences in the dependency of LPL for APOC2 with different types of lipid emulsions. For instance, synthetic lipid emulsions of triglycerides made up from long-chain fatty acids are hydrolyzed to some extent by LPL even without APOC2, referred to as the basal activity of LPL. Addition of APOC2 may increase the basal activity about 5-fold [166]. However, with APOC2-deficient chylomicrons as substrate, the basal activity of LPL was extremely low and it therefore increased >100-fold upon addition of exogenous APOC2 [167]. Structurally APOC2 is made up of three amphipathic α-helices all with negative polar faces spanning between residues 16–38, 45–57, and 65–74 respectively (Figure 5) [168,169]. Helix 1 and 2 are presumed to constitute the main lipid binding properties of APOC2 [170,171]. It was

20 concluded that the N-terminal APOC2 fragment (residues 1-50) could not activate LPL while the C-terminal fragment (residues 51-79) was responsible for activation of LPL [172]. It was shown that an APOC2 fragment spanning residues 50-79 could increase LPL activity against synthetic lipid emulsion particles, while full-length APOC2 was required when APOC2- deficient chylomicrons were used as substrate. The same authors suggested that the surface pressure for the individual lipid particles could account for these differences [167]. It was proposed by others that an APOC2 fragment (residues 39-62) could activate LPL [173], but these results could not be reproduced [174]. More recent studies confirmed the importance of helix 3 in LPL activation and indicate a direct interaction with the enzyme. In this study the nature of the fully conserved residues among animal species within the polar face of helix 3 were shown to be more critical for LPL activation than the hydrophobic residues on the opposite side of the helix [174].

There are no reports on successful generation of Apoc2-knockout mice. APOC2-transgenic mice develop hypertriglyceridemia attributed to delayed clearance of VLDL. Plasma APOC2 levels correlated positively with plasma triglycerides [175]. APOC2 was reported to impair the cellular uptake of remnant lipoproteins by either displacement of, or conformational modifications of, APOE [153]. However, it is also possible that APOC2-transgenic mice exhibit impaired intravascular lipolysis. Intercrossing APOC2-transgenic mice with muscle-specific LPL-transgenic mice reduced the hypertriglyceridemia associated with APOC2 overexpression [176]. This support the notion that excessive amounts of APOC2 inhibit LPL activity in vivo and suggest that LPL can compete with excessive APOC2 residing on TRLs.

Dogma holds that APOC2 is the LPL cofactor. The exact mechanism for how these proteins interact has not been resolved. For this, studies of the lipid-bound states for both APOC2 and LPL are necessary and steady state kinetics cannot be obtained for the required time due to lipid hydrolysis by LPL followed by remodeling of the lipid. Thus innovative approaches are required to explain how LPL is activated by APOC2.

Figure 5. High-resolution structure of human APOC2 in complex with dodecylphosphocholine micelles [177].

Apolipoprotein C3 Human subjects that are heterozygous for a premature stop mutation within the signal peptide R19X* in the APOC3 gene have reduced levels of plasma APOC3 compared to non- carriers. As a result, carriers display lower levels of plasma triglycerides and increased levels of HDL-cholesterol. In addition, carriers were shown to be less prone to develop coronary arterial calcification [178,179]. Other rare mutations in the APOC3 gene were reported to be associated with reduced levels of plasma APOC3, and with concomitant reductions in plasma triglyceride levels compared with controls; these include two splice site mutations [180], the A23T* and E58K* variants [181,182]. For the A23T and the splice site mutations, plasma APOB levels were not affected, suggesting more efficient intravascular lipolysis (for the E58K mutation no APOB measurements were reported). Recently it was shown that loss-of- function mutations in APOC3 are associated with protection against CVD [183,184].

* R19X refer to the position 19 with position one being the first amino acid residue of the signal peptide. A23T and E58K refer to the positions 23 and 58 respectively where position one being the first amino acid residue of the secreted protein.

22 APOC3 is a 79 amino acid residue glycoprotein. It is mainly synthesized by the liver, but is also synthesized in the intestine [164]. APOC3 is a constituent of chylomicrons, VLDL and HDL [6]. APOC3 inhibits LPL activity in vitro [185,186]. Alaupovic and coworkers concluded from studies on hypertriglyceridemic patients that plasma from these subjects had inhibitory effects on LPL activity and that this effect was due to APOC3. Using kinetic analysis the authors suggested that APOC3 inhibits LPL activity by a direct interaction with the enzyme [187]. A reduced degree of glycosylation of APOC3 was recently proposed to account for improved postprandial triglyceride clearance due to lowering the inhibition of APOC3 on LPL activity [188]. The authors also suggested that an altered distribution of APOC3 on plasma lipoproteins could account for their observation. From our unpublished observations using synthetic emulsions we could not see any difference between APOC3 glycosylation variants regarding inhibition of LPL activity. Thus, other explanations are needed. APOC3 is composed of 6 amphipathic helices with positively charged residues accessing the solvent for helices 1 and 2 while negatively charged residues comprise the polar face of the C-terminal helices 4 and 5 (Figure 6) [189]. Thrombin cleavage of APOC3 generates two fragments (residues 1–40 and 41–78), where the C-terminal fragment is mainly responsible for lipid binding and LPL inhibition [190,191]. For the rare natural mutation A23T in the APOC3 gene, the associated low levels of plasma triglycerides could not be explained by reduced inhibition of LPL compared to the wild-type protein when a synthetic lipid emulsion was used as substrate [182]. Moreover, the E58K mutation has not been shown to have impaired ability to inhibit LPL activity. Recently, it was proposed that APOC3 plays a crucial role in VLDL assembly, independently of MTP, and that both A23T and E58K were unable to incorporate bulk triglycerides into precursor lipoproteins [192,193]. APOC3 impairs APOE-mediated, but not

APOB100-mediated, uptake of remnant lipoproteins [153]. From studies of VLDL from hypertriglyceridemic subjects, Breyer et al. demonstrated that APOE was redistributed from VLDL to HDL at increasing APOC3 concentrations [194]. Interestingly, the redistribution of APOE was dependent on VLDL size, with large particles being less capable of ejecting APOE from the lipid/water interface.

Unlike Apoc1-knockout mice, Apoc3-knockout mice have decreased plasma cholesterol and triglyceride levels and do not display hypercholesterolemia when put on an atherogenic diet [195,196]. Combined Apoe/Apoc3-knockout mice display moderate reductions in plasma cholesterol and significant reductions in circulating triglyceride levels compared with Apoe- knockout mice. The postprandial response to an oral lipid load was blunted in both Apoc3- knockout mice and the combined Apoc3/Apoe-knockout mice. Particle clearance was not affected, suggesting profound effects on intravascular lipolysis. Chylomicron production rates were lower in combined knockout mice compared to Apoe-knockout mice, but the reductions in chylomicron secretion could not account for the dramatic effects seen on the postprandial lipid response [196].

Several species have been genetically modified to overexpress human APOC3 and in all cases the animals develop severe hypertriglyceridemia [197,198,199]. APOC3-transgenic mice

23 intercrossed with Apoe-knockout mice displayed increased plasma cholesterol and triglyceride levels without effects on lipoprotein production rates. Triglyceride levels were several-fold more increased than corresponding cholesterol levels suggesting that the main effect of APOC3 overexpression was due to inhibition of LPL activity [200].

Overexpression of APOC3 in mice causes a phenotype with high resemblance to that seen on overexpression of APOC1. Possibly all these observations can be explained by a redistribution of APOE from TRLs to other lipoproteins, causing impaired ability for the

APOC3-containing TRLs to interact with LPL, and prolonged residency for APOB48 remnant lipoproteins in blood. The redistribution of APOE might be a consequence of the excess of APOC3 that competes out APOE from TRLs. In support of this is that combined APOE- transgenic and APOC3-transgenic mice display normal levels of plasma triglycerides [201]. Additionally, APOC3 can displace APOC2 from lipid emulsion droplets [202]. Thus, it is conceivable that APOC3 (and APOC1) can displace APOC2 from TRLs and thereby impair lipolysis. In contrast, there are differences when Apoc1- and Apoc3-knockout mice are compared, either on a wild-type or an APOE-deficient, background suggesting distinct roles for these APOCs in lipoprotein metabolism.

Figure 6. High-resolution structure of human APOC3 in complex with sodium dodecyl sulfate micelles [189].

Apolipoprotein E APOE deficiency in humans is associated with premature CVD and manifests in moderately increased levels of plasma triglycerides and massive accumulation of cholesterol in remnant

24 lipoproteins due to impaired hepatic uptake via receptor-mediated endocytosis [203]. The human APOE gene is represented by three common alleles; APOE2, APOE3 and APOE4, where APOE3 is the normal allele with regard to all known functions. The APOE isoforms differ by having different combinations of arginine and/or cysteine residues at position 112 and 158, respectively. Allele-specific effects have been extensively studied in the fields of cardiovascular disease and alzheimers disease [reviewed [204]].

APOE is a ligand for members of the LDL-receptor family which internalize lipoproteins into endosomes and lysosomes of cells. The main function of APOE is to mediate uptake of remnant lipoproteins (chylomicron remnants and IDL) in the liver, mainly via LDL receptor- related protein 1 (LRP1) and LDLR. APOE is especially important for clearance of chylomicrons remnants because APOB48 cannot interact with members of the LDL-receptor family [12]. APOE is much more potent than APOB100 as ligand for LDLR, and may therefore regulate the residence time of lipoproteins in the circulation [204]. Unlike most other apolipoproteins, APOE is synthesized by numerous tissues. Large amounts are produced in the brain where APOE functions as a lipid transport vehicle in cerebrospinal fluid. APOE plays a central role in the neuronal response to injury by providing lipids which promote neuronal repair [204]. Macrophages secrete significant amounts of APOE. It was shown that APOE stimulate efflux of cholesterol from cholesterol-enriched macrophages to HDL in vitro [205,206].

Apoe-knockout mice show a similar phenotype as humans with dysfunctional APOE due to impaired removal of APOB48 remnant lipoproteins. In contrast to humans wild-type mice do not spontaneously develop atherosclerosis, while Apoe-knockout mice develop atherosclerotic plaques when put on a diet enriched in cholesterol [207]. Mice deficient of APOE, except for in macrophages, were protected against atherosclerosis when compared with Apoe-knockout mice [208]. Direct evidence for the importance of APOE for cholesterol- efflux from macrophages, and for prevention of atherosclerosis, was obtained by studies with bone-marrow transplantations. Transplantation of bone marrow from APOE deficient donor mice to wild-type mice showed that the recipient animals developed 10-fold more atherosclerosis than their littermates reconstituted with bone marrow from wild-type donors [209].

Apoe-knockout mice intercrossed with LPL-transgenic mice were protected against diet- induced atherosclerosis [87]. In this study, overexpression of LPL reduced plasma triglyceride levels while total cholesterol and APOB levels remained unchanged. These observations demonstrated that increased triglyceride hydrolysis of APOB48 remnant lipoproteins did not improve lipoprotein remnant particle removal from the circulation, nor did an excess of LPL under these conditions improve remnant uptake via cell surface heparan sulphate as previously proposed [210].

In in vitro experiments APOE inhibits LPL activity dose-dependently, both with lipid emulsion particles and APOE-deficient VLDL as substrate. Injection of lipid emulsion particles

25 containing recombinant APOE showed impaired lipolysis in hepatectomized rats dependent on the dose of APOE [211,212]. APOE-transgenic mice show moderately increased plasma triglyceride levels where plasma APOE levels correlate positively with plasma triglyceride levels and negatively with APOC2 on VLDL. This suggests that APOE may impair lipolysis, at least in part, by displacement of APOC2 from the TRLs. In addition, VLDL secretion was higher in APOE-transgenic mice. This could in part account for the observed hypertriglyceridemia [213]. As discussed above, APOE-transgenic mice ameliorate the hypertriglyceridemia associated with APOC3 overexpression. Thus, the inhibition of LPL activity by APOE in vivo warrants further studies.

Apolipoprotein A5

Homozygous nonsense mutations in the human gene of APOA5 are associated with severe hypertriglyceridemia and low levels of HDL-cholesterol [214,215]. Typically the hypertriglyceridemia is not as severe as that observed in homozygous loss-of-function mutations in any of the genes for LPL, GPIHBP1, APOC2 or LMF1.

APOA5 is mainly produced by the liver and associates with chylomicrons, VLDL and HDL [216,217]. Peculiar for this apolipoprotein are the low circulating levels in plasma ranging from 1 to 10 nmol/l [217]. In comparison, plasma levels of APOB100 and APOCs in fasted subjects were reported to be in the micromolar range [161,218]. The low plasma concentrations of APOA5 suggest intracellular functions for this protein. Ryan and coworkers have postulated that APOA5 promotes lipid droplet formation in hepatocytes and thereby modulate hepatic triglyceride stores by retarding VLDL assembly and subsequent secretion [219,220]. However, certain human mutations in the gene of APOA5 cause chylomicronemia suggesting effects also on circulating TRLs [214].

Transgenic APOA5-mice displayed reductions in plasma triglycerides without affecting cholesterol levels. Moreover, Apoa5-knockout mice displayed approximately 4-fold increased levels of plasma triglycerides, while cholesterol levels remained unaffected compared to wild-type mice [216]. This is a relatively mild phenotype compared to knockout mice models for other human mutations that cause severe hypertriglyceridemia (e.g. LPL, GPIHBP1, LMF1, APOC2). Possibly this is a species-dependent effect, but it could also indicate that other confounding factors are present in humans. Due to their opposing phenotypes; Apoc3-knockout mice were intercrossed with Apoa5-knockout mice. Similarly APOC3-transgenic mice were intercrossed with APOA5-transgenic mice. In both cases the resulting double knockout and double transgenic progeny displayed lipid levels within the normal range, compared to wild-type mice. Based on this the authors suggested that APOA5 and APOC3 affect plasma triglyceride levels independently but in an opposing manner [221].

26 Aims of the thesis

 To investigate the molecular mechanisms by which APOC3 inhibits LPL activity

 To compare the molecular mechanisms for inhibition of LPL by APOC1 and APOC3

 To investigate if fatty acids bind to ANGPTL4 and whether fatty acids affect the inactivation of LPL by ANGPTL4

 To conduct a small molecule screen designed to identify compounds that would block the ability of ANGPTL4 to inactivate LPL

 To perform structure-activity relationship studies on promising “hit” compounds to find lead compounds with improved potency to preserve LPL activity

27 Results and discussion

Paper I

In this study we present evidence for that APOC1 or APOC3, when bound to TRLs, prevents binding of LPL to the lipid/water interface of the substrate lipid droplets. This results in decreased lipolysis by the enzyme. In the presence of APOC2 more APOC1 or APOC3 was needed to prevent LPL from binding to lipid emulsion particles, and consequently to inhibit lipase activity. We found that prevention of LPL from binding to the lipid/water interface of TRLs and lipid emulsion particles rendered LPL more susceptible to inactivation by ANGPTL4. Mutagenesis of APOC3 revealed that alanine replacements within helix 3 and 4 of (W42A and F47A) were the most important changes that caused decreased lipid binding of APOC3 and consequently decreased inhibition of LPL activity.

Method development and considerations

To be able to assess binding of LPL to lipid emulsion particles, a filtration method was developed. By filtering the incubation mixtures through a syringe filter, LPL that was bound to lipid particles larger than the pore size of the filter would be retained. Previous methods had utilized the buoyant properties of triglyceride-rich particles to investigate LPL binding. The large triglyceride-rich lipid particles can be recovered from the top layer after centrifugation of the incubation mixtures and analyzed for their content of LPL [56,167]. However, lipolysis will go on during the time required for the centrifugation and this may impair the buoyant properties of the lipid particles. Another problem is that pressure effects will occur during the centrifugation [222]. Our technique avoids lipolysis by allowing for a very rapid separation of the incubation mixture. A disadvantage shared with the flotation techniques is that LPL bound to lipid particles smaller than the filter pore size are not distinguished from unbound LPL. This problem could probably be solved by using filters with an even smaller pore size. For centrifugation, the density of the mixtures could be increased to flotate even the smaller particles. However, the filtration method is quick and allows control regarding separation based on particles size. A potential disadvantage by using the filter approach is that unbound proteins may get stuck in the filter. Because of this we used filters composed of membranes with low protein-binding properties. We also used albumin in the incubation mixtures to block unspecific protein binding. In control experiments we observed >90% recovery of LPL from the separations, even in the absence of lipid particles, suggesting that binding of LPL to the filter was minimal.

Discussion

Previous studies, based on conventional analysis of enzyme kinetics, had proposed that APOC3 inhibits LPL by direct interaction with the enzyme at a site other than the active site [187]. However, such an interaction has not been possible to demonstrate. In contrast, our results suggest that APOC1 or APOC3 affects the properties of the substrate so that binding

28 of LPL to the lipid lipid/water interface is reduced. It is known that analyses of lipase kinetics are complicated due to that the lipid substrate is aggregated in emulsion droplets [223]. This makes the normal tests for enzyme inhibition non-applicable. A mechanism similar to ours was reported for an inhibitor of trypsin and thrombin, by which the inhibitor affected the substrates rather than interacted with the enzymes directly [224].

The exchangeable apolipoproteins differ in their ability to bind to and be retained on lipoproteins. It was demonstrated that when apolipoproteins penetrates a phospholipid monolayer the surface pressure increases as the interface becomes more crowded [225]. It was predicted that APOC1 has a higher affinity for lipoproteins than APOC2. Therefore it was speculated by others that binding of APOC1 to a TRL will result in a local increase of the surface pressure, leading to the displacement of APOC2 [151]. By using lipid emulsion particles carrying APOC2 it was shown that APOC3 inhibited LPL activity with a concomitant loss of APOC2 from the lipid/water interface [202]. We did not investigate if APOC1 or APOC3 displaced APOC2 in our systems, but in light of the previous report it is likely that this was the case also in our systems. In the absence of APOC2 the ability of LPL to hydrolyze triglycerides in TRLs is severely impaired [167], while with lipid emulsions containing long- chain triglycerides (e.g. Intralipid) LPL shows some triglyceride hydrolysis even in the absence of APOC2, suggesting that interaction between LPL and the surface lipids occur [56]. Our results suggest that APOC3 prevents this interaction, while APOC2 may promote such interactions. Based on our observations it should be expected that LPL is unable to bind to APOC2-deficient TRLs. This was, however, reported not to be the case with APOC2-deficient human TRLs that were isolated by flotation [167]. Further studies are needed to investigate these ambiguities that are likely due to that binding of both LPL and the apolipoproteins is highly dependent on the lipid composition and physical properties of the surface layer of the substrate emulsion particles, and that these properties may differ between synthetic emulsions and plasma lipoproteins.

Several studies have addressed the effects of APOCs on receptor-mediated endocytosis of lipoproteins. It was reported that addition of APOCs to VLDLs could displace APOE and that that this caused a reduction in cellular lipoprotein uptake. There are, however, conflicting opinions about whether the degree of APOE-displacement (APOE was reduced but still present) could account for the effects on lipoprotein uptake via the LRP1 [152,153,226]. This phenomenon could be due to displacement of LPL by APOCs, because it has been shown that LPL strongly enhance binding of APOE-containing lipoproteins to LRP1 [227]. Supporting that LPL could act as a ligand in vivo is that LPL is present in plasma at concentrations of about 100 ng/ml [228]. Interestingly, transgenic mice overexpressing LPL intercrossed with Apoe- knockout mice display low levels of plasma triglycerides, while circulating APOB levels are not affected compared to those in Apoe-knockout mice [87]. Possibly the combined presence of APOE and LPL on the remnant lipoproteins is necessary for receptor-mediated uptake.

29 A mysterious component in triglyceride metabolism is APOA5. It was suggested that APOA5 and APOC3 affect plasma triglyceride levels independently, but in an opposing manner [221]. In this study the combined APOC3/APOA5-trangenic mice overexpressed both proteins to the same magnitude, approximately 500-fold higher than their respective founder line. Given the observations that APOC3 can displace APOC2 and APOE, it is not unlikely that APOA5 could be displaced as well [194,202]. If APOC3 and APOA5 bind competitively to the lipid/water interface of TRLs, it is conceivable that the amount bound of each protein will depend on their respective concentrations and binding affinities. Thus, the observed effect that the combined transgenic mice displayed normal triglyceride levels was most likely dependent on the relative concentrations of the apolipoproteins on the surface of the TRLs. This cannot be considered as representing independent mechanisms for control of TRL metabolism. LPL-transgenic mice were able to normalize plasma triglycerides in Apoa5- knockout mice. It was suggested that APOA5 enhance intravascular lipolysis by guiding TRLs to proteoglycan-bound LPL [229]. This is an interesting hypothesis, especially in light of recent advances in triglyceride metabolism. As discussed above, several lines of evidence suggest that LPL is bound to its transporter protein GPIHBP1 also at the luminal side of the capillary endothelial cells. TRL margination in capillaries depends on LPL, but only when the enzyme is in complex with GPIHBP1 [129]. Thus, the hypothesis originally proposed by Merkel et al. might hold true, [229], with the updated modification that APOA5 via its affinity to GPIHBP1 could enhance LPL-dependent lipolysis by guiding TRLs to LPL anchored to the endothelial cells by GPIHBP1. APOA5 was shown to bind to GPIHBP1 in vitro. Later studies showed that recombinant APOA5 fails to reduce plasma triglyceride levels in Gpihbp1- knockout mice, but ameliorates much of the hypertriglyceridemia in Apoa5-knockout mice [76,230]. It should be noted, however, that Gpihbp1-knockout mice have extremely low amounts of LPL in capillaries. Taken together, these are interesting data that implicate an important role for APOA5 to enhance LPL-dependent lipolysis through interaction with GPIHBP1 and thereby tethering the TRLs to the endothelium.

We propose another consequence of the competition by APOC1 and APOC3 with LPL for binding to the surface of lipid particles, namely an increased susceptibility for LPL to become inactivated by ANGPTL4. Given that ANGPTL4 can readily inactivate LPL when the enzyme is free in solution, and that previous studies had shown that TRLs or lipid emulsion prevent this inactivation, we expected that APOC1 or APOC3 should promote inactivation of LPL by ANGPTL4 in the presence of lipid droplets. Our data in paper I show that this is in fact the case. Whether these effects occur also in vivo remain to be demonstrated. Transgenic mice that overexpressed APOC1 displayed increased levels of LPL activity in post-heparin plasma compared to wild-type mice [156]. With our proposed mechanism in mind this observation may be due to the inability of TRLs to remove LPL from the endothelium. Transgenic miniature pigs that overexpressed APOC3 did not show reduced post-heparin LPL activity after the effects of APOC3 were corrected for by serial dilution, while other studies demonstrated reduced LPL activity but had not taken this effect into account [199]. Thus, transgenic animal models that overexpress APOC1 or APOC3 do not indicate that LPL is more

30 susceptible to inactivation by ANGPTL4, but recombinant injection or overexpression of ANGPTL4 in mice reduces post-heparin LPL activity [97]. Later studies showed that this effect was reproducible in fasted mice, but interestingly not in littermates that had been given an olive oil bolus [106]. The same study suggested that the prime action of ANGPTL4 on LPL occurs by paracrine effects within interstitial spaces rather than at the capillary lumen. This was based on the knowledge about the low plasma concentrations of ANGPTL4 in humans and the protective effects on LPL activity seen by TRLs in blood. Our present observations could predict that under certain conditions, when plasma concentrations of APOC1 and/or APOC3 and ANGPTL4 are elevated, LPL activity could also be modulated at the capillary lumen.

Recently two independent studies have shown that rare heterozygous loss-of-function mutations in the gene of APOC3 are associated with reduced CVD risk [183,184]. Carriers show reductions in plasma triglyceride levels (-39%, -44%) and increased levels of HDL- cholesterol (+22%, +24%) compared to non-carriers. Plasma triglyceride levels were similar for all mutations which included two splice-site mutations, the nonsense mutation R19X and the missense mutation A23T. Interestingly, the A23T mutation is comparable to wild-type APOC3 for inhibition of LPL activity in incubation systems in vitro [182]. Therefore other mechanisms for the effect of APOC3 on plasma triglyceride levels must be considered. A recently proposed mechanism for the function of APOC3 was to assist in VLDL assembly and secretion. The A23T variant abolished the ability to stimulate VLDL secretion in vitro [192]. This is an interesting observation and challenges the preexisting view regarding the function of APOC3 for inhibition of intravascular lipolysis when this apolipoprotein is expressed at normal levels. It is possible to inhibit LPL activity on TRLs in blood by i.v. injections of triblock copolymers (e.g. triton wr-1339 or poloxamer 407) and thereby assess lipoprotein production rates [231]. In such experiments there were no differences in VLDL production rates between Apoc3-knockout mice compared to wild-type mice, suggesting that the main effect by APOC3 on plasma triglyceride levels is due to inhibition of LPL activity and/or impaired receptor mediated uptake [196]. Thus, more studies are needed to understand the mechanisms of the A23T mutation in vivo. The in vitro systems are not representative for the situation in blood which differs in two fundamental ways. Firstly, apolipoproteins can redistribute among the lipoprotein classes in blood. This is not possible in systems using lipid emulsions. It was shown that, compared to wild-type APOC3, the A23T mutation caused reduced affinity of LPL for phospholipid liposomes, while the inhibition of this mutant on LPL activity against emulsified triglycerides was not affected [182]. Mutants like the A23T may cause small changes of the properties of lipoproteins that reduce the affinity for LPL. These effects may not be detected in in vitro systems using lipid emulsions, whereas in blood they may cause an altered distribution of apolipoprotiens among lipoprotein classes, resulting in an impaired ability to inhibit LPL activity. Secondly, most mechanistic studies on apolipoproteins in relation to LPL activity have been conducted in test tube experiments with LPL free in solution. The recent findings that LPL is bound to GPIHBP1, also at its site of action in capillaries, give rise to a fundamental question; are TRLs more dependent on

31 apolipoproteins for efficient triglyceride hydrolysis when LPL is bound to GPIHBP1 compared to when LPL is free in solution? If this is the case, some known apolipoproteins may serve other functions than those we know of today. As discussed above, APOA5 might guide TRLs to LPL, and APOC2 may support binding of TRLs to the LPL-GPIHBP1 complex in addition to its direct enzyme activating properties.

In summary, I suggest that APOC1 and APOC3 retard TRL catabolism by preventing LPL from interaction with the lipid particles and by competition with other apolipoproteins for binding to the lipid/water interface. APOCs retard receptor-mediated uptake by displacement of APOE and my studies suggest that also LPL could be displaced. Further studies are needed to understand the details of the function of APOC1 and APOC3 in vivo by using in vitro systems that mimic the situation in capillaries where LPL is bound to GPIHBP1 and all lipoprotein classes are present.

Paper II

In this study we used surface plasmon resonance, isothermal titration calorimetry, and fluorescent quenching to investigate binding of fatty acids to ANGPTL4. It was shown that fatty acids bind with very high affinity to ANGPTL4 and that ANGPTL4, in the presence of fatty acids, has limited capacity to inactivate LPL. Addition of bovine serum albumin to incubation mixtures containing fatty acids, ANGPTL4 and LPL removed the protective effect of fatty acids on LPL activity, and dose-dependently promoted ANGPTL4-mediated inactivation of LPL.

Discussion

It was previously shown that LPL is inhibited by its products when incubated with long-chain triglyceride substrates, and that the inhibition is relieved by addition of a fatty acid acceptor like albumin [58]. In this study the authors proposed that the inhibition was mainly due to binding of fatty acids to LPL, and that the function might be to serve as a feed-back regulation to prevent excessive fatty acid delivery to cells. In the present study we found that fatty acids can bind even stronger to ANGPTL4 than to LPL. We also found that LPL is less prone to become inactivated by ANGPTL4 in the presence of fatty acids. Previous studies had demonstrated that TRLs protect LPL against ANGPTL4-mediated inactivation [106]. Possibly the fatty acids released during lipolysis of TRLs prevent ANGPTL4 from interacting with LPL. This may explain part of the observed protective effect. Because albumin was present during those incubations in vitro, and should sequester the fatty acids, it is also likely that binding of LPL to TRLs had protective effects on LPL activity.

In an in vivo situation it is difficult to reconcile the effects of fatty acids on LPL activity in adipose tissue in the fasted state when the expression of ANGPTL4 is up-regulated by activated PPARs, while the LPL system is tuned down and fatty acids are mobilized from intracellular lipid stores, presumably with the aid of ANGPTL4 [68,69]. The reduced ability of ANGPTL4 to inactivate LPL in the presence of fatty acids is not in accordance with the rapid

32 down-regulation of LPL activity during fasting [68]. Intriguingly, LPL activity in adipose tissue of obese individuals was shown to be unresponsive to a meal, while lean subjects displayed decreased specific LPL activity during fasting indicating inactivation of the enzyme [232]. Administration of the transcription inhibitor actinomycin D to old obese rats in order to block protein synthesis did not increase LPL activity in the fasted state as would have been expected due to decreased synthesis of ANGPTL4. This suggested impaired ability of ANGPTL4 to inactivate LPL in the obese animals [68]. In obesity, plasma free fatty acids are usually elevated due to insulin resistance, and hypoxia may occur in white adipose tissue [233,234]. Both factors may promote upregulation of ANGPTL4 expression. In human adipocytes hypoxia causes dramatic increases in ANGPTL4 expression [235]. Later studies have shown that in addition to hypoxia, fatty acids may elevate ANGPTL4 expression and secretion in human adipocytes [236]. Thus, it would be expected that adipose tissue LPL should be strongly suppressed in obesity. Evidently this is not the case. A possibility could be that the increased concentration of fatty acid that accompanies obesity neutralizes the ability of ANGPTL4 to inactivate LPL. More studies are obviously needed to understand the physiological relevance of the high affinity of ANGPTL4 for fatty acids.

Paper III and IV

In Paper III we have employed a small molecule screening approach to identify compounds that could protect LPL against inactivation by ANGPTL4. Paper IV is a follow-up structure- activity relationship (SAR) study aimed to develop lead compounds for drug development efforts for prevention of CVD. Small molecules that prevent inactivation of lipoprotein lipase (LPL) were identified. One hit compound (Figure 7) prevented both ANGPTL4-dependent and heat-dependent inactivation of LPL. This occurred by stabilization of the active LPL dimer. The hit compound reduced postprandial triglyceride levels in Apoa5-knockout mice. SAR studies resulted in a series of lead compounds, some of which showed approximately 2-3 fold improved potency compared to the original hit compound. SAR analysis identified the carboxyl-group (Figure 7, marked in orange) as highly important for retaining biological activity of the compound. Increasing the steric hindrance by substituent scrambling of the central phenyl ring (blue) decreased the compound potency and solubility. Modifications of the amide group (green) seemed tolerable. Hydrophilic substituents of either the piperidinyl group (yellow) or the para-tolyl group (red) reduced biological activity of the compound while more lipophilic substituents increased compound potency but decreased solubility.

Figure 7. Chemical structure of the hit compound used as starting point for our SAR investigation. The encapsulations denote chemical groups that were investigated by modification and/or substitution.

Although administration of our original hit compound showed promising lipid-lowering effects in Apoa5-knockout mice [Paper III], the phthalimide moiety (white) was shown to be metabolically unstable in the presence of human liver microsomes [Paper IV]. Primary metabolites were less potent than the original hit compound in vitro. Secondary metabolites did not show in vitro activity at 25 µM or below. For this reason, analogous heterocyclic compounds were synthesized to investigate if improved metabolic stability could increase compound efficacy in vivo. Although certain heterocyclic analogues showed improved potency in vitro the effects were moderate compared to the dramatic effects observed in vivo, suggesting that the original hit compound had a shorter half-life than the modified compounds in vivo (Table 2).

Original hit Hetero- Hetero- Hetero- Hetero- Hetero- Time Buffer compound cycle-A cycle-B cycle-C cycle-D cycle-E (min) (µg TG/ml) (µg TG/ml) (µg TG/ml) (µg TG/ml) (µg TG/ml) (µg TG/ml) (µg TG/ml)

0 349±112 325±104 263±97 271±76 217±73 196±75* 222±73

60 759±249 531±256 446±167** 297±114*** 379±161** 323±121*** 413±124**

120 921±241 802±377 507±110** 324±110*** 359±162*** 485±123** 521±283*

180 1043±656 1079±301 600±122 456±109** 421±96** 440±90** 436±159**

Table 2: In vivo effects of heterocyclic phthalimide substituents. Female C57B6J mice (n=56, n=8/group) aged 8-9 weeks with an average weight of 18 g were used for the in vivo experiments. The mice were housed at room temperature with free access to tap water and standard chow. The mice were treated in the morning for 3 days with 0.5 ml intraperitoneal injections (buffer composition as that described in Paper III containing 1 mM compound). On the 3rd day the mice had been fasted 8 h from midnight and injected a final 4th time. One hour after the final injection the mice were given an oral olive oil gavage of 0.2 ml. Blood samples were collected to EDTA coated capillaries from the tail vein at baseline and for every 60 min until 180 min. They were analyzed for their triglyceride contents. Data for blood triglyceride (TG) levels are represented as mean values with standard deviations. *p < 0.05, **p < 0.01, ***p<0.001.

34 Method development and considerations In the pharmaceutical industry high-throughput screening (HTS) is part of the early drug- discovery process. Once the decision has been made to try to affect the function of a target (e.g. protein of interest) the next step is to test if drug-like compounds possess biological activity towards the target. Typically more than 100000 compounds are screened in fully automated facilities. Therefore it is necessary to develop robust assays that are amenable to automation. Usually traditional benchtop assays for a given target are transformed to a microtiter plate format to enable robotic pipetting and facilitate endpoint readouts. Assay reliability is easily assessed by applying various statistics based on controls (e.g. no compound) or the total signal distribution [237]. Reliable assays are able to identify biologically active compounds, usually referred to as “hits”, while discarding inactive ones. In the screening process it is most common to use one concentration at one time and then cherry-pick hits for re-screening by using the same assay. Hits are further validated by secondary assays and by dose-response studies to better assess if a hit compound interacts with the target, or if the effects are due to off-target effects. Using a compound library of approximately 17000 compounds these steps were carried out in the facilities of Laboratories for Chemical Biology Umeå and form the basis for Paper III. Compounds that remain after validation studies may be used as starting points for lead optimization studies. In the lead optimization step SAR studies are used to study the relationship between a compounds molecular structure and its biological activity. By the identification and substitution of functional groups and structurally relevant properties, these studies aim to increase the potency of the original hit compounds. A lead optimization was carried out for our hit compound and is described in Paper IV.

Traditional benchtop assays for measuring LPL activity include triglyceride-based substrates (e.g. TRLs or lipid emulsions) and water-soluble ester substrates (e.g. para-nitrophenyl butyrate). When triglyceride substrates are used, a subsequent method for quantification of fatty acids is required. Quantification of radiolabelled fatty acids is most commonly used in in vitro assays for LPL. This procedure is not easily transformed to an automated format. An alternative quantification method for fatty acid is titration using a pH-stat, but this is too laborious in a screening environment. An attractive but expensive option was to use quantification of non-labelled fatty acids by an enzymatic calorimetric assay. To obtain an affordable, yet reliable, assay we used a water-soluble substrate for LPL in our primary screen. LPL has catalytic activity to both emulsified triglyceride substrates and water-soluble ester substrates. However, previous studies had shown that on inactivation of LPL the enzymatic activity disappears more quickly towards long-chain triglyceride emulsions than towards water-soluble substrates [238]. This is presumably because the water-soluble substrates do not require binding of LPL to the lipid/water interface of emulsion particles [55]. Thus, by using water-soluble esters as substrates we expected that we should not miss out on potential hits. Rather, we took the risk of identifying compounds that showed effects on LPL with water-soluble ester substrates, but not with emulsified triglyceride substrates.

35 However, we reasoned that such compounds could be identified and later discarded by using a long-chain triglyceride-based substrate in a secondary screen. We first used the water- soluble para-nitrophenyl butyrate as substrate [239]. However, numerous compounds appeared to catalyze hydrolysis of this substrate themselves, leading to large numbers of false positives. We therefore turned to an umbelliferone-ester (pivaloyl-umbelliferone) which had shown better resistance to non-specific hydrolysis [240]. There were yet no reports regarding whether this ester could function as a substrate for LPL. Interestingly, in comparison with para-nitrophenyl butyrate, the hydrolysis of the umbelliferone-ester was linear over a very long time suggesting that LPL was stabilized by the substrate and/or by the product formed. Addition of ANGPTL4 caused marked reductions in product formation. We therefore decided to use this substrate in our primary screen. Our second concern was to keep LPL stable during the automated process. One problem in this regard was that traditional stabilizers of LPL activity (e.g. bile salts or fatty acids) did not enable ANGPTL4 to inactivate LPL. Other known stabilizers, such as heparin and similar polyanions, are less efficient and bind LPL by ionic interactions that could prevent interactions with compounds. For these reasons we decided to use conditions for preserving LPL activity without the addition of stabilizers. It appeared that LPL was relatively stable in phosphate buffer for activity against pivaloyl-umbelliferone for sessions up to three hours. Our screening campaign had a throughput of approximately 600 compounds per hour. Given the relatively small compound library, we did try to increase throughput further. Validation and follow-up studies are described in detail in Paper III.

Discussion

The idea of finding a drug that prevents inactivation of LPL by ANGPTL4 in order to reduce the risk for CVD is based on the knowledge that increased levels of LPL activity is associated with an anti-atherogenic lipid profile and that carriers of the LPL-S447X gain-of-function mutation have been associated with reduced CVD risk [31,41,42]. Also, the E40K mutation of ANGPTL4, that has impaired ability to inactivate LPL in vitro, is associated with an anti- atherogenic lipid profile [112,113]. However, there are conflicting results regarding the effects on CVD risk for this ANGPTL4 mutation [115,116]. Transgenic mice overexpressing LPL at locations other than in macrophages are protected against atherosclerosis, while overexpression of LPL in macrophages is pro-atherogenic and was found to mainly be due to the non-catalytic functions of LPL [93]. An alarming fact is that when Angptl4-knockout mice are fed a diet enriched in saturated fat they develop lipid-laden macrophages in the mesenteric lymph nodes [101]. It was also shown that ANGPTL4 suppresses LPL activity in macrophages and that this may reduce atherosclerotic lesion progression [110,111]. Given these contradictory findings for the role of LPL with regard to presumed CVD risk it is impossible to speculate on the outcomes for a drug that would increase LPL activity in vivo. Our goal was, however, to find a drug that can reduce the levels of plasma triglycerides. The obvious target for this effort is LPL. Irrespective of outcome, a compound that directly affects LPL activity is an interesting tool for future studies.

36 Recently, another group reported on a small molecule as a novel LPL agonist [241]. They used para-nitrophenyl butyrate as substrate for LPL and ANGPTL4 for inactivation, an approach almost identical to that of our first screening effort. Similar to several compounds in our primary screen, the published LPL agonist failed to show activity in our secondary screen using triglyceride-based substrates [Paper III]. The reason for this discrepancy is currently unknown, but as discussed in Paper III it might be due to lack of stabilization of the necessary lipid-binding properties of LPL. Other drugs that may affect LPL activity are those which prevent the hypertriglyceridemia induced by triblock co-polymers (e.g. triton wr- 1339). It should be noted that the mechanism by which triblock co-polymers inhibit LPL is not understood. One compound class which prevented triton wr-1339 induced hypertriglyceridemia showed structural similarities to our compounds [Paper IV,[242]]. However, this compound class did not show activity in our assays. Furthermore administration of triton wr-1339 to mice receiving our original hit compound showed a similar rise in plasma triglyceride levels as mice receiving vehicle only. The drug Ibrolipim (NO-1886) was reported to increase post-heparin LPL activity, and to decrease plasma triglycerides, increase HDL-cholesterol and protect against atherosclerosis in rats and rabbits [243,244]. The exact mechanism by which Ibrolipim increases LPL activity has not been elucidated. Both LPL mRNA levels and LPL mass is increased upon Ibrolipim administration, suggesting transcriptional effects [245]. Ibrolipim protect against diet-induced atherosclerosis in miniature pigs [246]. The same animals displayed increased post-heparin LPL activity, but they also had increased levels of ABCA1 mRNA and mass, presumably due to increased levels and activation of the liver X receptor alpha (LXRα) [246]. Activation of LXR has many effects and protects against atherosclerosis [reviewed [247]]. This makes it hard to discriminate the exact mode of action by which Ibrolipim may protect against atherosclerosis.

Small molecules with biological activity in our assays show a common property; they are all anionic amphiphilic compounds. This property is shared with fatty acids and is also a property in common with deoxycholic acid that is known to efficiently stabilize LPL in vitro [Paper II][248]. We have observed that deoxycholic acid prevents inactivation of LPL by ANGPTL4 (unpublished data). Both fatty acids and deoxycholic acid function as ligands for nuclear receptors involved in lipid metabolism where fatty acids primarily activate PPARs, while deoxycholic acid activates the farnesoid X receptor [249,250,251]. Thus, it is highly important to investigate if our compounds show off-target effects on nuclear receptors and if these effects may account for our observations regarding plasma lipids in vivo. Both LPL and ANGPTL4 bind with high affinity to fatty acids as well as to our original hit compound [Paper II, Paper III]. It is therefore likely that our lead compounds bind to both proteins as well.

With regard to the mode of action of ANGPTL4 on LPL an appealing idea has emerged during our studies. It is evident that LPL and ANGPTL4 show similar physical properties with regard to hydrophobic and charged sites. It was previously shown that rapid subunit exchange

37 occurs in the active LPL dimer [238]. Thus, it is possible that ANGPTL4 mimics the structure of an LPL monomer, and once a transiently free LPL monomer encounters ANGPTL4 they may bind to each other. The suboptimal partner ANGPTL4 may not be able to stabilize the conformation of the LPL subunit as well as another LPL subunit. Therefore the interaction with ANGPTL4 results in irreversible inactivation of LPL. The presence of anionic amphiphilic compounds may shift the equilibrium in favour of the dimeric state of LPL and thereby prevent the enzyme from inactivation by ANGPTL4 (Figure 8). Another possibility is that the anionic amphiphiles simply prevent interaction between ANGPTL4 and LPL by blocking the binding site.

Figure 8. Model for stabilization of the active LPL dimer by compounds and ANGPTL4-mediated inactivation involving binding of ANGPTL4 to dissociated, but still natively folded LPL monomers.

In terms of the lipid-lowering effects of our compounds, preliminary in vivo data are highly encouraging. Administration of our stable heterocyclic lead compounds to wild-type mice caused dramatic reductions in plasma triglycerides. Studies are ongoing to validate these compounds regarding off-target effects and to assess if they are effective in the prevention of atherosclerosis in animal models.

38 Conclusions

 APOC1 and APOC3 inhibit LPL activity by similar mechanisms. Both proteins prevent LPL from binding to the lipid-water interface of TRLs and lipid emulsion particles.

 Hydrophobic amino acid residues centrally located in APOC3 are the most important residues for lipid-binding, and consequently for inhibition of LPL activity.

 ANGPTL4 and LPL both have high affinity for fatty acids. More studies are needed to understand the physiological relevance of this interaction.

 Anionic amphiphilic small molecules stabilize the LPL homodimer and prevent inactivation of LPL by ANGPTL4.

 Heterocyclic substitution of our original hit compound caused dramatic improvements in plasma lipid parameters in vivo, presumably by improved metabolic stability of the compound.

39 Acknowledgements

Gunilla tack för att du alltid tagit dig tid att lyssna på mina tankar kring forskning. Du har gett mig den frihet och vägledning jag behövde för att få utvecklas och komma till insikt med vad jag vill göra i min fortsatta yrkeskarriär. För detta är jag mycket tacksam.

Nästa tack går till alla medarbetare på fysiologisk kemi. Jag uppskattar verkligen all hjälp, ert sällskap och trevliga stunder. Aivar tack, för all hjälp och trevliga diskussioner. Jag hoppas fortfarande komma till Tallinn och hälsa på. Solveig, en riddare utan rustning. Såväl röd som svart är hon den vänligaste av oss alla. Elena, ”I will kill you” har fått en helt ny innebörd. Numera associerar jag det med omtänksamma människor och nektariner. Lasse har ett problem – han glömmer inte – vare sig allmäna kunskaper eller var han inte har lagt sitt kaffekort. Stort tack för tillfällen då ett lyssnande öra behövts. Thomas is still going strong, en sann forskare. Stefan, hade vänligheten att bjuda in mig till tjockisfältet, olyckligtvis blev jag tunnare. ”Krabban” a.k.a byggar-Rakel, Mållgan hälsar att han inte behöver fler imaginära buffertar. Evelina har ingenting emot LPL aktivitet, vilket antyder att hon måste ha det starkaste psyket i vintergatan. Madde, vi ses på six flags! Fredrik och Niklas det känns tryggt att lämna över till er, kom ihåg att låten ”Ai Se Eu Te Pego” kan ge nektariner. Oleg, en blivande sann forskare. Slava, Jessica och Massi thank you for many laughs at the end.

Ett stort tack till tjejerna i tarmen på våning två. Öppna dörr nummer ett och du tas emot och blir servad på det bästa av vis av en alltid så glad Åsa. Innanför dörr nummer två huserar ”aldrig ett problem”-Carina, där man tyvärr aldrig behöver vistas så länge. Bakom dörr nummer tre sliter vår hjälte Terry såväl sena kvällar som helger. Längst in, finner vi en annan slitvarg och även den bästa av bordsdamer Karin.

Ett alldeles eget tack tillägnas Clara som även hon har sett till att mitt arbete har förflutit sömlöst.

En eloge till tjejerna på pediatrik som stått ut med surgubben i labbet intill. Yvonne, jag håller på dig! Sussi stay frosty. Anna fortsätt att le. Carina du skrattar högst, fortsätt med det. Catarina slår du hammaren såsom ELISA-plattan tycker jag synd om plankan. Lotta till skillnad från solen har du inga fläckar.

Stort tack till mina vänner på organisk kemi. Remi your diligence and professionalism inspires me. Mikael för att du ställer upp i tid och otid, uppskattas! P-A, en vänlig själ.

Tack även till alla er som gjort 6M till en bra och trivsam arbetsplats. Maria B dig skulle man klona! Tack för all hjälp. Ingegärd tack för trevliga stunder och ugnsbak. Maria L som ställer upp och sekvenserar fastän du inte behöver. Monica och Maria H blir det något lopp? Lelle för att du muntrar upp. Malin för att du är så okonstlat bra. Angelica för att du är kul. Nina för att du fått mig att försöka vara en pistvakt och Urban för att du bistått i mina försök. Åsa kom och ät. Lisbeth, Mikael, Heidi för att ni alltid är glada. Emma för bästa humorn. Linda

40 för att jag får försöka programmera. Sofia för att du är snäll. Per som delar min passion för litterära mästerverk. Dan, Elin och Martin, jag hoppas vi ses igen.

Cecilia Elofsson, en räddare i nöden. Tack!

Till sist, tack till mina föräldrar och min bror. En annan tid, ett annat liv–jag önskar vi får vara tillsammans även då.

41 References

[1] G.L. Zubay, Biochemistry, 4th ed., Wm.C. Brown Publishers, Dubuque, IA, 1998. [2] M.E. Lowe, Pancreatic triglyceride lipase and colipase: insights into dietary fat digestion, Gastroenterology 107 (1994) 1524-1536. [3] M.M. Hussain, Intestinal lipid absorption and lipoprotein formation, Curr Opin Lipidol 25 (2014) 200-206. [4] D.I. Abramson, P.B. Dobrin, Blood vessels and lymphatics in organ systems, Academic Press, Orlando, 1984. [5] I. Ramasamy, Recent advances in physiological lipoprotein metabolism, Clin Chem Lab Med (2013) 1-33. [6] R.J. Havel, J.P. Kane, M.L. Kashyap, Interchange of apolipoproteins between chylomicrons and high density lipoproteins during alimentary lipemia in man, J Clin Invest 52 (1973) 32-38. [7] S.G. Young, R. Zechner, Biochemistry and pathophysiology of intravascular and intracellular lipolysis, Genes Dev 27 (2013) 459-484. [8] P.J. Nestel, High-density lipoprotein turnover, Am Heart J 113 (1987) 518-521. [9] A.D. Cooper, Hepatic uptake of chylomicron remnants, J Lipid Res 38 (1997) 2173-2192. [10] S. Tiwari, S.A. Siddiqi, Intracellular trafficking and secretion of VLDL, Arterioscler Thromb Vasc Biol 32 (2012) 1079-1086. [11] S.J. Murdoch, W.C. Breckenridge, Effect of lipid transfer proteins on lipoprotein lipase induced transformation of VLDL and HDL, Biochim Biophys Acta 1303 (1996) 222-232. [12] R.W. Mahley, Z.S. Ji, Remnant lipoprotein metabolism: key pathways involving cell-surface heparan sulfate proteoglycans and apolipoprotein E, J Lipid Res 40 (1999) 1-16. [13] A. Zambon, S. Bertocco, N. Vitturi, V. Polentarutti, D. Vianello, G. Crepaldi, Relevance of hepatic lipase to the metabolism of triacylglycerol-rich lipoproteins, Biochem Soc Trans 31 (2003) 1070-1074. [14] R.W. Mahley, Apolipoprotein E: cholesterol transport protein with expanding role in cell biology, Science 240 (1988) 622-630. [15] M.S. Brown, J.L. Goldstein, A receptor-mediated pathway for cholesterol homeostasis, Science 232 (1986) 34-47. [16] J.L. Goldstein, M.S. Brown, The LDL receptor, Arterioscler Thromb Vasc Biol 29 (2009) 431-438. [17] M. Cofan Pujol, [Basic mechanisms: absorption and excretion of cholesterol and other sterols], Clin Investig Arterioscler 26 (2014) 41-47. [18] R.S. Kiss, D.C. McManus, V. Franklin, W.L. Tan, A. McKenzie, G. Chimini, Y.L. Marcel, The lipidation by hepatocytes of human apolipoprotein A-I occurs by both ABCA1-dependent and -independent pathways, J Biol Chem 278 (2003) 10119-10127. [19] J. Huuskonen, V.M. Olkkonen, M. Jauhiainen, C. Ehnholm, The impact of phospholipid transfer protein (PLTP) on HDL metabolism, Atherosclerosis 155 (2001) 269-281. [20] G. Li, H.M. Gu, D.W. Zhang, ATP-binding cassette transporters and cholesterol translocation, IUBMB Life 65 (2013) 505-512. [21] J.A. Glomset, E.T. Janssen, R. Kennedy, J. Dobbins, Role of plasma lecithin:cholesterol acyltransferase in the metabolism of high density lipoproteins, J Lipid Res 7 (1966) 638-648. [22] A. Leiva, H. Verdejo, M.L. Benitez, A. Martinez, D. Busso, A. Rigotti, Mechanisms regulating hepatic SR-BI expression and their impact on HDL metabolism, Atherosclerosis 217 (2011) 299-307. [23] T. Chajek, C.J. Fielding, Isolation and characterization of a human serum cholesteryl ester transfer protein, Proc Natl Acad Sci U S A 75 (1978) 3445-3449. [24] C.J. O'Donnell, E.G. Nabel, Genomics of cardiovascular disease, N Engl J Med 365 (2011) 2098- 2109. [25] G.K. Hansson, Inflammation, atherosclerosis, and coronary artery disease, N Engl J Med 352 (2005) 1685-1695.

42 [26] I. Tabas, K.J. Williams, J. Boren, Subendothelial lipoprotein retention as the initiating process in atherosclerosis: update and therapeutic implications, Circulation 116 (2007) 1832-1844. [27] C. Baigent, L. Blackwell, J. Emberson, L.E. Holland, C. Reith, N. Bhala, R. Peto, E.H. Barnes, A. Keech, J. Simes, R. Collins, Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials, Lancet 376 (2010) 1670-1681. [28] M. Miller, N.J. Stone, C. Ballantyne, V. Bittner, M.H. Criqui, H.N. Ginsberg, A.C. Goldberg, W.J. Howard, M.S. Jacobson, P.M. Kris-Etherton, T.A. Lennie, M. Levi, T. Mazzone, S. Pennathur, Triglycerides and cardiovascular disease: a scientific statement from the American Heart Association, Circulation 123 (2011) 2292-2333. [29] A.M. Gotto, Jr., J.E. Moon, Pharmacotherapies for lipid modification: beyond the statins, Nat Rev Cardiol 10 (2013) 560-570. [30] J.P. Despres, HDL cholesterol studies--more of the same?, Nat Rev Cardiol 10 (2013) 70-72. [31] C.T. Johansen, R.A. Hegele, The complex genetic basis of plasma triglycerides, Curr Atheroscler Rep 14 (2012) 227-234. [32] T. Gotoda, K. Shirai, T. Ohta, J. Kobayashi, S. Yokoyama, S. Oikawa, H. Bujo, S. Ishibashi, H. Arai, S. Yamashita, M. Harada-Shiba, M. Eto, T. Hayashi, H. Sone, H. Suzuki, N. Yamada, Diagnosis and management of type I and type V hyperlipoproteinemia, J Atheroscler Thromb 19 (2012) 1-12. [33] C.T. Johansen, S. Kathiresan, R.A. Hegele, Genetic determinants of plasma triglycerides, J Lipid Res 52 (2011) 189-206. [34] G.G. Schwartz, A.G. Olsson, M. Abt, C.M. Ballantyne, P.J. Barter, J. Brumm, B.R. Chaitman, I.M. Holme, D. Kallend, L.A. Leiter, E. Leitersdorf, J.J. McMurray, H. Mundl, S.J. Nicholls, P.K. Shah, J.C. Tardif, R.S. Wright, Effects of dalcetrapib in patients with a recent acute coronary syndrome, N Engl J Med 367 (2012) 2089-2099. [35] W.E. Boden, J.L. Probstfield, T. Anderson, B.R. Chaitman, P. Desvignes-Nickens, K. Koprowicz, R. McBride, K. Teo, W. Weintraub, Niacin in patients with low HDL cholesterol levels receiving intensive statin therapy, N Engl J Med 365 (2011) 2255-2267. [36] M. Jun, C. Foote, J. Lv, B. Neal, A. Patel, S.J. Nicholls, D.E. Grobbee, A. Cass, J. Chalmers, V. Perkovic, Effects of fibrates on cardiovascular outcomes: a systematic review and meta- analysis, Lancet 375 (2010) 1875-1884. [37] P.F. Hahn, Abolishment of Alimentary Lipemia Following Injection of Heparin, Science 98 (1943) 19-20. [38] E.D. Korn, Clearing factor, a heparin-activated lipoprotein lipase. I. Isolation and characterization of the enzyme from normal rat heart, J Biol Chem 215 (1955) 1-14. [39] V. Murthy, P. Julien, C. Gagne, Molecular pathobiology of the human lipoprotein lipase gene, Pharmacol Ther 70 (1996) 101-135. [40] M. Merkel, R.H. Eckel, I.J. Goldberg, Lipoprotein lipase: genetics, lipid uptake, and regulation, J Lipid Res 43 (2002) 1997-2006. [41] M.K. Jensen, E.B. Rimm, D. Rader, E.B. Schmidt, T.I. Sorensen, U. Vogel, K. Overvad, K.J. Mukamal, S447X variant of the lipoprotein lipase gene, lipids, and risk of coronary heart disease in 3 prospective cohort studies, Am Heart J 157 (2009) 384-390. [42] J. Rip, M.C. Nierman, C.J. Ross, J.W. Jukema, M.R. Hayden, J.J. Kastelein, E.S. Stroes, J.A. Kuivenhoven, Lipoprotein lipase S447X: a naturally occurring gain-of-function mutation, Arterioscler Thromb Vasc Biol 26 (2006) 1236-1245. [43] H. Wong, M.C. Schotz, The lipase gene family, J Lipid Res 43 (2002) 993-999. [44] H. van Tilbeurgh, A. Roussel, J.M. Lalouel, C. Cambillau, Lipoprotein lipase. Molecular model based on the pancreatic lipase x-ray structure: consequences for heparin binding and catalysis, J Biol Chem 269 (1994) 4626-4633. [45] H. Wong, R.C. Davis, T. Thuren, J.W. Goers, J. Nikazy, M. Waite, M.C. Schotz, Lipoprotein lipase domain function, J Biol Chem 269 (1994) 10319-10323.

43 [46] J.C. Osborne, Jr., G. Bengtsson-Olivecrona, N.S. Lee, T. Olivecrona, Studies on inactivation of lipoprotein lipase: role of the dimer to monomer dissociation, Biochemistry 24 (1985) 5606- 5611. [47] H. Wong, R.C. Davis, J.S. Hill, D. Yang, M.C. Schotz, Lipase engineering: a window into structure- function relationships, Methods Enzymol 284 (1997) 171-184. [48] Y. Kobayashi, T. Nakajima, I. Inoue, Molecular modeling of the dimeric structure of human lipoprotein lipase and functional studies of the carboxyl-terminal domain, Eur J Biochem 269 (2002) 4701-4710. [49] G. Bengtsson-Olivecrona, T. Olivecrona, Phospholipase activity of milk lipoprotein lipase, Methods Enzymol 197 (1991) 345-356. [50] G. Bengtsson-Olivecrona, T. Olivecrona, Binding of active and inactive forms of lipoprotein lipase to heparin. Effects of pH, Biochem J 226 (1985) 409-413. [51] K. Arnold, L. Bordoli, J. Kopp, T. Schwede, The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling, Bioinformatics 22 (2006) 195-201. [52] L. Bordoli, F. Kiefer, K. Arnold, P. Benkert, J. Battey, T. Schwede, Protein structure homology modeling using SWISS-MODEL workspace, Nat Protoc 4 (2009) 1-13. [53] H. van Tilbeurgh, L. Sarda, R. Verger, C. Cambillau, Structure of the pancreatic lipase-procolipase complex, Nature 359 (1992) 159-162. [54] T.O. Olivecrona, G., Cellular Lipid Metabolism, Springer Berlin Heidelberg, 2009. [55] A. Lookene, G. Bengtsson-Olivecrona, Chymotryptic cleavage of lipoprotein lipase. Identification of cleavage sites and functional studies of the truncated molecule, Eur J Biochem 213 (1993) 185-194. [56] G. Bengtsson, T. Olivecrona, Lipoprotein lipase: some effects of activator proteins, Eur J Biochem 106 (1980) 549-555. [57] J.S. Hill, D. Yang, J. Nikazy, L.K. Curtiss, J.T. Sparrow, H. Wong, Subdomain chimeras of hepatic lipase and lipoprotein lipase. Localization of heparin and cofactor binding, J Biol Chem 273 (1998) 30979-30984. [58] G. Bengtsson, T. Olivecrona, Lipoprotein lipase. Mechanism of product inhibition, Eur J Biochem 106 (1980) 557-562. [59] C.F. Semenkovich, C.C. Luo, M.K. Nakanishi, S.H. Chen, L.C. Smith, L. Chan, In vitro expression and site-specific mutagenesis of the cloned human lipoprotein lipase gene. Potential N-linked glycosylation site asparagine 43 is important for both enzyme activity and secretion, J Biol Chem 265 (1990) 5429-5433. [60] R. Busca, M.A. Pujana, P. Pognonec, J. Auwerx, S.S. Deeb, M. Reina, S. Vilaro, Absence of N- glycosylation at asparagine 43 in human lipoprotein lipase induces its accumulation in the rough endoplasmic reticulum and alters this cellular compartment, J Lipid Res 36 (1995) 939- 951. [61] M. Peterfy, O. Ben-Zeev, H.Z. Mao, D. Weissglas-Volkov, B.E. Aouizerat, C.R. Pullinger, P.H. Frost, J.P. Kane, M.J. Malloy, K. Reue, P. Pajukanta, M.H. Doolittle, Mutations in LMF1 cause combined lipase deficiency and severe hypertriglyceridemia, Nat Genet 39 (2007) 1483-1487. [62] O. Ben-Zeev, M. Hosseini, C.M. Lai, N. Ehrhardt, H. Wong, A.B. Cefalu, D. Noto, M.R. Averna, M.H. Doolittle, M. Peterfy, Lipase maturation factor 1 is required for endothelial lipase activity, J Lipid Res 52 (2011) 1162-1169. [63] P.H. Weinstock, C.L. Bisgaier, K. Aalto-Setala, H. Radner, R. Ramakrishnan, S. Levak-Frank, A.D. Essenburg, R. Zechner, J.L. Breslow, Severe hypertriglyceridemia, reduced high density lipoprotein, and neonatal death in lipoprotein lipase knockout mice. Mild hypertriglyceridemia with impaired very low density lipoprotein clearance in heterozygotes, J Clin Invest 96 (1995) 2555-2568. [64] K.J. Excoffon, G. Liu, L. Miao, J.E. Wilson, B.M. McManus, C.F. Semenkovich, T. Coleman, P. Benoit, N. Duverger, D. Branellec, P. Denefle, M.R. Hayden, M.E. Lewis, Correction of hypertriglyceridemia and impaired fat tolerance in lipoprotein lipase-deficient mice by

44 adenovirus-mediated expression of human lipoprotein lipase, Arterioscler Thromb Vasc Biol 17 (1997) 2532-2539. [65] M. Shimada, H. Shimano, T. Gotoda, K. Yamamoto, M. Kawamura, T. Inaba, Y. Yazaki, N. Yamada, Overexpression of human lipoprotein lipase in transgenic mice. Resistance to diet-induced hypertriglyceridemia and hypercholesterolemia, J Biol Chem 268 (1993) 17924-17929. [66] P. Benlian, Genetics of dyslipidemia, Kluwer Academic Publishers, Boston, 2001. [67] M. Bergo, G. Olivecrona, T. Olivecrona, Forms of lipoprotein lipase in rat tissues: in adipose tissue the proportion of inactive lipase increases on fasting, Biochem J 313 ( Pt 3) (1996) 893- 898. [68] M. Bergo, G. Wu, T. Ruge, T. Olivecrona, Down-regulation of adipose tissue lipoprotein lipase during fasting requires that a gene, separate from the lipase gene, is switched on, J Biol Chem 277 (2002) 11927-11932. [69] V. Sukonina, A. Lookene, T. Olivecrona, G. Olivecrona, Angiopoietin-like protein 4 converts lipoprotein lipase to inactive monomers and modulates lipase activity in adipose tissue, Proc Natl Acad Sci U S A 103 (2006) 17450-17455. [70] W. Dijk, S. Kersten, Regulation of lipoprotein lipase by Angptl4, Trends Endocrinol Metab 25 (2014) 146-155. [71] S. Kersten, Physiological regulation of lipoprotein lipase, Biochim Biophys Acta 1841 (2014) 919- 933. [72] T. Ruge, M. Bergo, M. Hultin, G. Olivecrona, T. Olivecrona, Nutritional regulation of binding sites for lipoprotein lipase in rat heart, Am J Physiol Endocrinol Metab 278 (2000) E211-218. [73] G.Q. Liu, T. Olivecrona, Pulse-chase study on lipoprotein lipase in perfused guinea pig heart, Am J Physiol 261 (1991) H2044-2050. [74] G. Wu, G. Olivecrona, T. Olivecrona, The distribution of lipoprotein lipase in rat adipose tissue. Changes with nutritional state engage the extracellular enzyme, J Biol Chem 278 (2003) 11925-11930. [75] G. Wu, L. Zhang, J. Gupta, G. Olivecrona, T. Olivecrona, A transcription-dependent mechanism, akin to that in adipose tissue, modulates lipoprotein lipase activity in rat heart, Am J Physiol Endocrinol Metab 293 (2007) E908-915. [76] A.P. Beigneux, B.S. Davies, P. Gin, M.M. Weinstein, E. Farber, X. Qiao, F. Peale, S. Bunting, R.L. Walzem, J.S. Wong, W.S. Blaner, Z.M. Ding, K. Melford, N. Wongsiriroj, X. Shu, F. de Sauvage, R.O. Ryan, L.G. Fong, A. Bensadoun, S.G. Young, Glycosylphosphatidylinositol-anchored high- density lipoprotein-binding protein 1 plays a critical role in the lipolytic processing of chylomicrons, Cell Metab 5 (2007) 279-291. [77] L. Bey, M.T. Hamilton, Suppression of skeletal muscle lipoprotein lipase activity during physical inactivity: a molecular reason to maintain daily low-intensity activity, J Physiol 551 (2003) 673-682. [78] D.B. Zilversmit, A proposal linking atherogenesis to the interaction of endothelial lipoprotein lipase with triglyceride-rich lipoproteins, Circ Res 33 (1973) 633-638. [79] B.G. Nordestgaard, D.B. Zilversmit, Large lipoproteins are excluded from the arterial wall in diabetic cholesterol-fed rabbits, J Lipid Res 29 (1988) 1491-1500. [80] J.R. Mead, A. Cryer, D.P. Ramji, Lipoprotein lipase, a key role in atherosclerosis?, FEBS Lett 462 (1999) 1-6. [81] M.M. Weinstein, L. Yin, Y. Tu, X. Wang, X. Wu, L.W. Castellani, R.L. Walzem, A.J. Lusis, L.G. Fong, A.P. Beigneux, S.G. Young, Chylomicronemia elicits atherosclerosis in mice--brief report, Arterioscler Thromb Vasc Biol 30 (2010) 20-23. [82] X. Zhang, R. Qi, X. Xian, F. Yang, M. Blackstein, X. Deng, J. Fan, C. Ross, J. Karasinska, M.R. Hayden, G. Liu, Spontaneous atherosclerosis in aged lipoprotein lipase-deficient mice with severe hypertriglyceridemia on a normal chow diet, Circ Res 102 (2008) 250-256. [83] L. Eiselein, D.W. Wilson, M.W. Lame, J.C. Rutledge, Lipolysis products from triglyceride-rich lipoproteins increase endothelial permeability, perturb zonula occludens-1 and F-actin, and induce apoptosis, Am J Physiol Heart Circ Physiol 292 (2007) H2745-2753.

45 [84] J.C. Rutledge, A.E. Mullick, G. Gardner, I.J. Goldberg, Direct visualization of lipid deposition and reverse lipid transport in a perfused artery : roles of VLDL and HDL, Circ Res 86 (2000) 768- 773. [85] F. Karpe, J.R. Dickmann, K.N. Frayn, Fatty acids, obesity, and insulin resistance: time for a reevaluation, Diabetes 60 (2011) 2441-2449. [86] M. Shimada, S. Ishibashi, T. Inaba, H. Yagyu, K. Harada, J.I. Osuga, K. Ohashi, Y. Yazaki, N. Yamada, Suppression of diet-induced atherosclerosis in low density lipoprotein receptor knockout mice overexpressing lipoprotein lipase, Proc Natl Acad Sci U S A 93 (1996) 7242- 7246. [87] H. Yagyu, S. Ishibashi, Z. Chen, J. Osuga, M. Okazaki, S. Perrey, T. Kitamine, M. Shimada, K. Ohashi, K. Harada, F. Shionoiri, N. Yahagi, T. Gotoda, Y. Yazaki, N. Yamada, Overexpressed lipoprotein lipase protects against atherosclerosis in apolipoprotein E knockout mice, J Lipid Res 40 (1999) 1677-1685. [88] J. Fan, H. Unoki, N. Kojima, H. Sun, H. Shimoyamada, H. Deng, M. Okazaki, H. Shikama, N. Yamada, T. Watanabe, Overexpression of lipoprotein lipase in transgenic rabbits inhibits diet- induced hypercholesterolemia and atherosclerosis, J Biol Chem 276 (2001) 40071-40079. [89] T. Ichikawa, S. Kitajima, J. Liang, T. Koike, X. Wang, H. Sun, M. Okazaki, M. Morimoto, H. Shikama, T. Watanabe, N. Yamada, J. Fan, Overexpression of lipoprotein lipase in transgenic rabbits leads to increased small dense LDL in plasma and promotes atherosclerosis, Lab Invest 84 (2004) 715-726. [90] V.R. Babaev, M.B. Patel, C.F. Semenkovich, S. Fazio, M.F. Linton, Macrophage lipoprotein lipase promotes foam cell formation and atherosclerosis in low density lipoprotein receptor- deficient mice, J Biol Chem 275 (2000) 26293-26299. [91] V.R. Babaev, S. Fazio, L.A. Gleaves, K.J. Carter, C.F. Semenkovich, M.F. Linton, Macrophage lipoprotein lipase promotes foam cell formation and atherosclerosis in vivo, J Clin Invest 103 (1999) 1697-1705. [92] M. Van Eck, R. Zimmermann, P.H. Groot, R. Zechner, T.J. Van Berkel, Role of macrophage-derived lipoprotein lipase in lipoprotein metabolism and atherosclerosis, Arterioscler Thromb Vasc Biol 20 (2000) E53-62. [93] M. Gustafsson, M. Levin, K. Skalen, J. Perman, V. Friden, P. Jirholt, S.O. Olofsson, S. Fazio, M.F. Linton, C.F. Semenkovich, G. Olivecrona, J. Boren, Retention of low-density lipoprotein in atherosclerotic lesions of the mouse: evidence for a role of lipoprotein lipase, Circ Res 101 (2007) 777-783. [94] X. Wu, J. Wang, J. Fan, M. Chen, L. Chen, W. Huang, G. Liu, Localized vessel expression of lipoprotein lipase in rabbits leads to rapid lipid deposition in the balloon-injured arterial wall, Atherosclerosis 187 (2006) 65-73. [95] S.C. Rumsey, J.C. Obunike, Y. Arad, R.J. Deckelbaum, I.J. Goldberg, Lipoprotein lipase-mediated uptake and degradation of low density lipoproteins by fibroblasts and macrophages, J Clin Invest 90 (1992) 1504-1512. [96] I. Tabas, Y. Li, R.W. Brocia, S.W. Xu, T.L. Swenson, K.J. Williams, Lipoprotein lipase and sphingomyelinase synergistically enhance the association of atherogenic lipoproteins with smooth muscle cells and extracellular matrix. A possible mechanism for low density lipoprotein and lipoprotein(a) retention and macrophage foam cell formation, J Biol Chem 268 (1993) 20419-20432. [97] A. Koster, Y.B. Chao, M. Mosior, A. Ford, P.A. Gonzalez-DeWhitt, J.E. Hale, D. Li, Y. Qiu, C.C. Fraser, D.D. Yang, J.G. Heuer, S.R. Jaskunas, P. Eacho, Transgenic angiopoietin-like (angptl)4 overexpression and targeted disruption of angptl4 and angptl3: regulation of triglyceride metabolism, Endocrinology 146 (2005) 4943-4950. [98] S. Kersten, S. Mandard, N.S. Tan, P. Escher, D. Metzger, P. Chambon, F.J. Gonzalez, B. Desvergne, W. Wahli, Characterization of the fasting-induced adipose factor FIAF, a novel peroxisome proliferator-activated receptor target gene, J Biol Chem 275 (2000) 28488-28493.

46 [99] H. Ge, J.Y. Cha, H. Gopal, C. Harp, X. Yu, J.J. Repa, C. Li, Differential regulation and properties of angiopoietin-like proteins 3 and 4, J Lipid Res 46 (2005) 1484-1490. [100] F. Mattijssen, S. Alex, H.J. Swarts, A.K. Groen, E.M. van Schothorst, S. Kersten, Angptl4 serves as an endogenous inhibitor of intestinal lipid digestion, Mol Metab 3 (2014) 135-144. [101] L. Lichtenstein, F. Mattijssen, N.J. de Wit, A. Georgiadi, G.J. Hooiveld, R. van der Meer, Y. He, L. Qi, A. Koster, J.T. Tamsma, N.S. Tan, M. Muller, S. Kersten, Angptl4 protects against severe proinflammatory effects of saturated fat by inhibiting fatty acid uptake into mesenteric lymph node macrophages, Cell Metab 12 (2010) 580-592. [102] A. Georgiadi, L. Lichtenstein, T. Degenhardt, M.V. Boekschoten, M. van Bilsen, B. Desvergne, M. Muller, S. Kersten, Induction of cardiac Angptl4 by dietary fatty acids is mediated by peroxisome proliferator-activated receptor beta/delta and protects against fatty acid- induced oxidative stress, Circ Res 106 (2010) 1712-1721. [103] K. Yoshida, T. Shimizugawa, M. Ono, H. Furukawa, Angiopoietin-like protein 4 is a potent hyperlipidemia-inducing factor in mice and inhibitor of lipoprotein lipase, J Lipid Res 43 (2002) 1770-1772. [104] L. Lichtenstein, J.F. Berbee, S.J. van Dijk, K.W. van Dijk, A. Bensadoun, I.P. Kema, P.J. Voshol, M. Muller, P.C. Rensen, S. Kersten, Angptl4 upregulates cholesterol synthesis in liver via inhibition of LPL- and HL-dependent hepatic cholesterol uptake, Arterioscler Thromb Vasc Biol 27 (2007) 2420-2427. [105] M.J. Lafferty, K.C. Bradford, D.A. Erie, S.B. Neher, Angiopoietin-like protein 4 inhibition of lipoprotein lipase: evidence for reversible complex formation, J Biol Chem 288 (2013) 28524- 28534. [106] S.K. Nilsson, F. Anderson, M. Ericsson, M. Larsson, E. Makoveichuk, A. Lookene, J. Heeren, G. Olivecrona, Triacylglycerol-rich lipoproteins protect lipoprotein lipase from inactivation by ANGPTL3 and ANGPTL4, Biochim Biophys Acta 1821 (2012) 1370-1378. [107] U. Desai, E.C. Lee, K. Chung, C. Gao, J. Gay, B. Key, G. Hansen, D. Machajewski, K.A. Platt, A.T. Sands, M. Schneider, I. Van Sligtenhorst, A. Suwanichkul, P. Vogel, N. Wilganowski, J. Wingert, B.P. Zambrowicz, G. Landes, D.R. Powell, Lipid-lowering effects of anti-angiopoietin- like 4 antibody recapitulate the lipid phenotype found in angiopoietin-like 4 knockout mice, Proc Natl Acad Sci U S A 104 (2007) 11766-11771. [108] H. Adachi, Y. Fujiwara, T. Kondo, T. Nishikawa, R. Ogawa, T. Matsumura, N. Ishii, R. Nagai, K. Miyata, M. Tabata, H. Motoshima, N. Furukawa, K. Tsuruzoe, J. Kawashima, M. Takeya, S. Yamashita, G.Y. Koh, A. Nagy, T. Suda, Y. Oike, E. Araki, Angptl 4 deficiency improves lipid metabolism, suppresses foam cell formation and protects against atherosclerosis, Biochem Biophys Res Commun 379 (2009) 806-811. [109] A. Lookene, N. Skottova, G. Olivecrona, Interactions of lipoprotein lipase with the active-site inhibitor tetrahydrolipstatin (Orlistat), Eur J Biochem 222 (1994) 395-403. [110] A. Georgiadi, Y. Wang, R. Stienstra, N. Tjeerdema, A. Janssen, A. Stalenhoef, J.A. van der Vliet, A. de Roos, J.T. Tamsma, J.W. Smit, N.S. Tan, M. Muller, P.C. Rensen, S. Kersten, Overexpression of angiopoietin-like protein 4 protects against atherosclerosis development, Arterioscler Thromb Vasc Biol 33 (2013) 1529-1537. [111] E. Makoveichuk, V. Sukonina, O. Kroupa, P. Thulin, E. Ehrenborg, T. Olivecrona, G. Olivecrona, Inactivation of lipoprotein lipase occurs on the surface of THP-1 macrophages where oligomers of angiopoietin-like protein 4 are formed, Biochem Biophys Res Commun 425 (2012) 138-143. [112] S. Romeo, L.A. Pennacchio, Y. Fu, E. Boerwinkle, A. Tybjaerg-Hansen, H.H. Hobbs, J.C. Cohen, Population-based resequencing of ANGPTL4 uncovers variations that reduce triglycerides and increase HDL, Nat Genet 39 (2007) 513-516. [113] V. Sukonina, Angiopoietin-like protein 4: an unfolding chaperone regulating lipoprotein lipase activity, Diss., Medical Biosciences, Umeå University, Umeå, 2007, pp. 50.

47 [114] S. Romeo, W. Yin, J. Kozlitina, L.A. Pennacchio, E. Boerwinkle, H.H. Hobbs, J.C. Cohen, Rare loss- of-function mutations in ANGPTL family members contribute to plasma triglyceride levels in humans, J Clin Invest 119 (2009) 70-79. [115] P.J. Talmud, M. Smart, E. Presswood, J.A. Cooper, V. Nicaud, F. Drenos, J. Palmen, M.G. Marmot, S.M. Boekholdt, N.J. Wareham, K.T. Khaw, M. Kumari, S.E. Humphries, ANGPTL4 E40K and T266M: effects on plasma triglyceride and HDL levels, postprandial responses, and CHD risk, Arterioscler Thromb Vasc Biol 28 (2008) 2319-2325. [116] A.R. Folsom, J.M. Peacock, E. Demerath, E. Boerwinkle, Variation in ANGPTL4 and risk of coronary heart disease: the Atherosclerosis Risk in Communities Study, Metabolism 57 (2008) 1591-1596. [117] P. Zhu, Y.Y. Goh, H.F. Chin, S. Kersten, N.S. Tan, Angiopoietin-like 4: a decade of research, Biosci Rep 32 (2012) 211-219. [118] R. Koishi, Y. Ando, M. Ono, M. Shimamura, H. Yasumo, T. Fujiwara, H. Horikoshi, H. Furukawa, Angptl3 regulates lipid metabolism in mice, Nat Genet 30 (2002) 151-157. [119] I. Minicocci, A. Montali, M.R. Robciuc, F. Quagliarini, V. Censi, G. Labbadia, C. Gabiati, G. Pigna, M.L. Sepe, F. Pannozzo, D. Lutjohann, S. Fazio, M. Jauhiainen, C. Ehnholm, M. Arca, Mutations in the ANGPTL3 gene and familial combined hypolipidemia: a clinical and biochemical characterization, J Clin Endocrinol Metab 97 (2012) E1266-1275. [120] K. Fujimoto, R. Koishi, T. Shimizugawa, Y. Ando, Angptl3-null mice show low plasma lipid concentrations by enhanced lipoprotein lipase activity, Exp Anim 55 (2006) 27-34. [121] M. Shimamura, M. Matsuda, H. Yasumo, M. Okazaki, K. Fujimoto, K. Kono, T. Shimizugawa, Y. Ando, R. Koishi, T. Kohama, N. Sakai, K. Kotani, R. Komuro, T. Ishida, K. Hirata, S. Yamashita, H. Furukawa, I. Shimomura, Angiopoietin-like protein3 regulates plasma HDL cholesterol through suppression of endothelial lipase, Arterioscler Thromb Vasc Biol 27 (2007) 366-372. [122] W.K. Sonnenburg, D. Yu, E.C. Lee, W. Xiong, G. Gololobov, B. Key, J. Gay, N. Wilganowski, Y. Hu, S. Zhao, M. Schneider, Z.M. Ding, B.P. Zambrowicz, G. Landes, D.R. Powell, U. Desai, GPIHBP1 stabilizes lipoprotein lipase and prevents its inhibition by angiopoietin-like 3 and angiopoietin-like 4, J Lipid Res 50 (2009) 2421-2429. [123] L. Shan, X.C. Yu, Z. Liu, Y. Hu, L.T. Sturgis, M.L. Miranda, Q. Liu, The angiopoietin-like proteins ANGPTL3 and ANGPTL4 inhibit lipoprotein lipase activity through distinct mechanisms, J Biol Chem 284 (2009) 1419-1424. [124] J. Liu, H. Afroza, D.J. Rader, W. Jin, Angiopoietin-like protein 3 inhibits lipoprotein lipase activity through enhancing its cleavage by proprotein convertases, J Biol Chem 285 (2010) 27561- 27570. [125] X. Lei, F. Shi, D. Basu, A. Huq, S. Routhier, R. Day, W. Jin, Proteolytic processing of angiopoietin- like protein 4 by proprotein convertases modulates its inhibitory effects on lipoprotein lipase activity, J Biol Chem 286 (2011) 15747-15756. [126] L. Zhang, A. Lookene, G. Wu, G. Olivecrona, Calcium triggers folding of lipoprotein lipase into active dimers, J Biol Chem 280 (2005) 42580-42591. [127] F. Quagliarini, Y. Wang, J. Kozlitina, N.V. Grishin, R. Hyde, E. Boerwinkle, D.M. Valenzuela, A.J. Murphy, J.C. Cohen, H.H. Hobbs, Atypical angiopoietin-like protein that regulates ANGPTL3, Proc Natl Acad Sci U S A 109 (2012) 19751-19756. [128] B.S. Davies, A.P. Beigneux, R.H. Barnes, 2nd, Y. Tu, P. Gin, M.M. Weinstein, C. Nobumori, R. Nyren, I. Goldberg, G. Olivecrona, A. Bensadoun, S.G. Young, L.G. Fong, GPIHBP1 is responsible for the entry of lipoprotein lipase into capillaries, Cell Metab 12 (2010) 42-52. [129] C.N. Goulbourne, P. Gin, A. Tatar, C. Nobumori, A. Hoenger, H.B. Jiang, C.R.M. Grovenor, O. Adeyo, J.D. Esko, I.J. Goldberg, K. Reue, P. Tontonoz, A. Bensadoun, A.P. Beigneux, S.G. Young, L.G. Fong, The GPIHBP1-LPL Complex Is Responsible for the Margination of Triglyceride-Rich Lipoproteins in Capillaries, Cell Metab 19 (2014) 849-860. [130] O. Adeyo, C.N. Goulbourne, A. Bensadoun, A.P. Beigneux, L.G. Fong, S.G. Young, Glycosylphosphatidylinositol-anchored high-density lipoprotein-binding protein 1 and the intravascular processing of triglyceride-rich lipoproteins, J Intern Med 272 (2012) 528-540.

48 [131] R. Franssen, S.G. Young, F. Peelman, J. Hertecant, J.A. Sierts, A.W. Schimmel, A. Bensadoun, J.J. Kastelein, L.G. Fong, G.M. Dallinga-Thie, A.P. Beigneux, Chylomicronemia with low postheparin lipoprotein lipase levels in the setting of GPIHBP1 defects, Circ Cardiovasc Genet 3 (2010) 169-178. [132] A.P. Beigneux, GPIHBP1 and the processing of triglyceride-rich lipoproteins, Clin Lipidol 5 (2010) 575-582. [133] A.P. Beigneux, P. Gin, B.S. Davies, M.M. Weinstein, A. Bensadoun, L.G. Fong, S.G. Young, Highly conserved cysteines within the Ly6 domain of GPIHBP1 are crucial for the binding of lipoprotein lipase, J Biol Chem 284 (2009) 30240-30247. [134] M.M. Weinstein, L. Yin, A.P. Beigneux, B.S. Davies, P. Gin, K. Estrada, K. Melford, J.R. Bishop, J.D. Esko, G.M. Dallinga-Thie, L.G. Fong, A. Bensadoun, S.G. Young, Abnormal patterns of lipoprotein lipase release into the plasma in GPIHBP1-deficient mice, J Biol Chem 283 (2008) 34511-34518. [135] J.P. Segrest, M.K. Jones, H. De Loof, C.G. Brouillette, Y.V. Venkatachalapathi, G.M. Anantharamaiah, The amphipathic helix in the exchangeable apolipoproteins: a review of secondary structure and function, J Lipid Res 33 (1992) 141-166. [136] J.P. Segrest, R.L. Jackson, J.D. Morrisett, A.M. Gotto, Jr., A molecular theory of lipid-protein interactions in the plasma lipoproteins, FEBS Lett 38 (1974) 247-258. [137] J.P. Segrest, H. De Loof, J.G. Dohlman, C.G. Brouillette, G.M. Anantharamaiah, Amphipathic helix motif: classes and properties, Proteins 8 (1990) 103-117. [138] W.H. Li, M. Tanimura, C.C. Luo, S. Datta, L. Chan, The apolipoprotein multigene family: biosynthesis, structure, structure-function relationships, and evolution, J Lipid Res 29 (1988) 245-271. [139] C.C. Luo, W.H. Li, M.N. Moore, L. Chan, Structure and evolution of the apolipoprotein multigene family, J Mol Biol 187 (1986) 325-340. [140] E.J. Schaefer, R.D. Santos, B.F. Asztalos, Marked HDL deficiency and premature coronary heart disease, Curr Opin Lipidol 21 (2010) 289-297. [141] M.C. Cheung, S.D. Sibley, J.P. Palmer, J.F. Oram, J.D. Brunzell, Lipoprotein lipase and hepatic lipase: their relationship with HDL subspecies Lp(A-I) and Lp(A-I,A-II), J Lipid Res 44 (2003) 1552-1558. [142] A.R. Tall, P.H. Green, R.M. Glickman, J.W. Riley, Metabolic fate of chylomicron phospholipids and apoproteins in the rat, J Clin Invest 64 (1979) 977-989. [143] S. Kaser, A. Sandhofer, B. Holzl, R. Gander, C.F. Ebenbichler, B. Paulweber, J.R. Patsch, Phospholipid and cholesteryl ester transfer are increased in lipoprotein lipase deficiency, J Intern Med 253 (2003) 208-216. [144] A.K. Soutar, C.W. Garner, H.N. Baker, J.T. Sparrow, R.L. Jackson, A.M. Gotto, L.C. Smith, Effect of the human plasma apolipoproteins and phosphatidylcholine acyl donor on the activity of lecithin: cholesterol acyltransferase, Biochemistry 14 (1975) 3057-3064. [145] S. Acton, A. Rigotti, K.T. Landschulz, S. Xu, H.H. Hobbs, M. Krieger, Identification of scavenger receptor SR-BI as a high density lipoprotein receptor, Science 271 (1996) 518-520. [146] S.G. Young, Recent progress in understanding apolipoprotein B, Circulation 82 (1990) 1574- 1594. [147] J. Greeve, I. Altkemper, J.H. Dieterich, H. Greten, E. Windler, Apolipoprotein B mRNA editing in 12 different mammalian species: hepatic expression is reflected in low concentrations of apoB-containing plasma lipoproteins, J Lipid Res 34 (1993) 1367-1383. [148] S.J. Lauer, D. Walker, N.A. Elshourbagy, C.A. Reardon, B. Levy-Wilson, J.M. Taylor, Two copies of the human apolipoprotein C-I gene are linked closely to the apolipoprotein E gene, J Biol Chem 263 (1988) 7277-7286. [149] R.J. Havel, C.J. Fielding, T. Olivecrona, V.G. Shore, P.E. Fielding, T. Egelrud, Cofactor activity of protein components of human very low density lipoproteins in the hydrolysis of triglycerides by lipoproteins lipase from different sources, Biochemistry 12 (1973) 1828-1833.

49 [150] A. Rozek, J.T. Sparrow, K.H. Weisgraber, R.J. Cushley, Conformation of human apolipoprotein C- I in a lipid-mimetic environment determined by CD and NMR spectroscopy, Biochemistry 38 (1999) 14475-14484. [151] N.L. Meyers, L. Wang, O. Gursky, D.M. Small, Changes in helical content or net charge of apolipoprotein C-I alter its affinity for lipid/water interfaces, J Lipid Res 54 (2013) 1927-1938. [152] K.H. Weisgraber, R.W. Mahley, R.C. Kowal, J. Herz, J.L. Goldstein, M.S. Brown, Apolipoprotein C- I modulates the interaction of apolipoprotein E with beta-migrating very low density lipoproteins (beta-VLDL) and inhibits binding of beta-VLDL to low density lipoprotein receptor-related protein, J Biol Chem 265 (1990) 22453-22459. [153] E. Sehayek, S. Eisenberg, Mechanisms of inhibition by apolipoprotein C of apolipoprotein E- dependent cellular metabolism of human triglyceride-rich lipoproteins through the low density lipoprotein receptor pathway, J Biol Chem 266 (1991) 18259-18267. [154] J.H. van Ree, M.H. Hofker, W.J. van den Broek, J.M. van Deursen, H. van der Boom, R.R. Frants, B. Wieringa, L.M. Havekes, Increased response to cholesterol feeding in apolipoprotein C1- deficient mice, Biochem J 305 ( Pt 3) (1995) 905-911. [155] N.S. Shachter, T. Ebara, R. Ramakrishnan, G. Steiner, J.L. Breslow, H.N. Ginsberg, J.D. Smith, Combined hyperlipidemia in transgenic mice overexpressing human apolipoprotein Cl, J Clin Invest 98 (1996) 846-855. [156] J.F. Berbee, C.C. van der Hoogt, D. Sundararaman, L.M. Havekes, P.C. Rensen, Severe hypertriglyceridemia in human APOC1 transgenic mice is caused by apoC-I-induced inhibition of LPL, J Lipid Res 46 (2005) 297-306. [157] M. Westerterp, W. de Haan, J.F. Berbee, L.M. Havekes, P.C. Rensen, Endogenous apoC-I increases hyperlipidemia in apoE-knockout mice by stimulating VLDL production and inhibiting LPL, J Lipid Res 47 (2006) 1203-1211. [158] T. Gautier, D. Masson, J.P. de Barros, A. Athias, P. Gambert, D. Aunis, M.H. Metz-Boutigue, L. Lagrost, Human apolipoprotein C-I accounts for the ability of plasma high density lipoproteins to inhibit the cholesteryl ester transfer protein activity, J Biol Chem 275 (2000) 37504-37509. [159] W.C. Breckenridge, J.A. Little, G. Steiner, A. Chow, M. Poapst, Hypertriglyceridemia associated with deficiency of apolipoprotein C-II, N Engl J Med 298 (1978) 1265-1273. [160] R. Fellin, G. Baggio, A. Poli, J. Augustin, M.R. Baiocchi, G. Baldo, M. Sinigaglia, H. Greten, G. Crepaldi, Familial lipoprotein lipase and apolipoprotein C-II deficiency. Lipoprotein and apoprotein analysis, adipose tissue and hepatic lipoprotein lipase levels in seven patients and their first degree relatives, Atherosclerosis 49 (1983) 55-68. [161] M.C. Jong, M.H. Hofker, L.M. Havekes, Role of ApoCs in lipoprotein metabolism: functional differences between ApoC1, ApoC2, and ApoC3, Arterioscler Thromb Vasc Biol 19 (1999) 472-484. [162] A.A. Kei, T.D. Filippatos, V. Tsimihodimos, M.S. Elisaf, A review of the role of apolipoprotein C-II in lipoprotein metabolism and cardiovascular disease, Metabolism 61 (2012) 906-921. [163] P. Fornengo, A. Bruno, R. Gambino, M. Cassader, G. Pagano, Resistant hypertriglyceridemia in a patient with high plasma levels of apolipoprotein CII, Arterioscler Thromb Vasc Biol 20 (2000) 2329-2339. [164] V.I. Zannis, F.S. Cole, C.L. Jackson, D.M. Kurnit, S.K. Karathanasis, Distribution of apolipoprotein A-I, C-II, C-III, and E mRNA in fetal human tissues. Time-dependent induction of apolipoprotein E mRNA by cultures of human monocyte-macrophages, Biochemistry 24 (1985) 4450-4455. [165] J.C. LaRosa, R.I. Levy, P. Herbert, S.E. Lux, D.S. Fredrickson, A specific apoprotein activator for lipoprotein lipase, Biochem Biophys Res Commun 41 (1970) 57-62. [166] Y. Andersson, A. Lookene, Y. Shen, S. Nilsson, L. Thelander, G. Olivecrona, Guinea pig apolipoprotein C-II: expression in E. coli, functional studies of recombinant wild-type and mutated variants, and distribution on plasma lipoproteins, J Lipid Res 38 (1997) 2111-2124.

50 [167] G. Olivecrona, U. Beisiegel, Lipid binding of apolipoprotein CII is required for stimulation of lipoprotein lipase activity against apolipoprotein CII-deficient chylomicrons, Arterioscler Thromb Vasc Biol 17 (1997) 1545-1549. [168] C.A. MacRaild, D.M. Hatters, G.J. Howlett, P.R. Gooley, NMR structure of human apolipoprotein C-II in the presence of sodium dodecyl sulfate, Biochemistry 40 (2001) 5414-5421. [169] J. Zdunek, G.V. Martinez, J. Schleucher, P.O. Lycksell, Y. Yin, S. Nilsson, Y. Shen, G. Olivecrona, S. Wijmenga, Global structure and dynamics of human apolipoprotein CII in complex with micelles: evidence for increased mobility of the helix involved in the activation of lipoprotein lipase, Biochemistry 42 (2003) 1872-1889. [170] C.E. MacPhee, G.J. Howlett, W.H. Sawyer, Mass spectrometry to characterize the binding of a peptide to a lipid surface, Anal Biochem 275 (1999) 22-29. [171] M. Dahim, W.E. Momsen, M.M. Momsen, H.L. Brockman, Specificity of the lipid-binding domain of apoC-II for the substrates and products of lipolysis, J Lipid Res 42 (2001) 553-562. [172] T.A. Musliner, E.C. Church, P.N. Herbert, M.J. Kingston, R.S. Shulman, Lipoprotein lipase cofactor activity of a carboxyl-terminal peptide of apolipoprotein C-II, Proc Natl Acad Sci U S A 74 (1977) 5358-5362. [173] C.E. MacPhee, D.M. Hatters, W.H. Sawyer, G.J. Howlett, Apolipoprotein C-II39-62 activates lipoprotein lipase by direct lipid-independent binding, Biochemistry 39 (2000) 3433-3440. [174] Y. Shen, A. Lookene, L. Zhang, G. Olivecrona, Site-directed mutagenesis of apolipoprotein CII to probe the role of its secondary structure for activation of lipoprotein lipase, J Biol Chem 285 (2010) 7484-7492. [175] N.S. Shachter, T. Hayek, T. Leff, J.D. Smith, D.W. Rosenberg, A. Walsh, R. Ramakrishnan, I.J. Goldberg, H.N. Ginsberg, J.L. Breslow, Overexpression of apolipoprotein CII causes hypertriglyceridemia in transgenic mice, J Clin Invest 93 (1994) 1683-1690. [176] L.K. Pulawa, D.R. Jensen, A. Coates, R.H. Eckel, Reduction of plasma triglycerides in apolipoprotein C-II transgenic mice overexpressing lipoprotein lipase in muscle, J Lipid Res 48 (2007) 145-151. [177] C.A. MacRaild, G.J. Howlett, P.R. Gooley, The structure and interactions of human apolipoprotein C-II in dodecyl phosphocholine, Biochemistry 43 (2004) 8084-8093. [178] T.I. Pollin, C.M. Damcott, H. Shen, S.H. Ott, J. Shelton, R.B. Horenstein, W. Post, J.C. McLenithan, L.F. Bielak, P.A. Peyser, B.D. Mitchell, M. Miller, J.R. O'Connell, A.R. Shuldiner, A null mutation in human APOC3 confers a favorable plasma lipid profile and apparent cardioprotection, Science 322 (2008) 1702-1705. [179] I. Tachmazidou, G. Dedoussis, L. Southam, A.E. Farmaki, G.R. Ritchie, D.K. Xifara, A. Matchan, K. Hatzikotoulas, N.W. Rayner, Y. Chen, T.I. Pollin, J.R. O'Connell, L.M. Yerges-Armstrong, C. Kiagiadaki, K. Panoutsopoulou, J. Schwartzentruber, L. Moutsianas, E. Tsafantakis, C. Tyler- Smith, G. McVean, Y. Xue, E. Zeggini, A rare functional cardioprotective APOC3 variant has risen in frequency in distinct population isolates, Nat Commun 4 (2013) 2872. [180] A.E. Bochem, J.C. van Capelleveen, G.M. Dallinga-Thie, A.W. Schimmel, M.M. Motazacker, I. Tietjen, R.R. Singaraja, M.R. Hayden, J.J. Kastelein, E.S. Stroes, G.K. Hovingh, Two novel mutations in apolipoprotein C3 underlie atheroprotective lipid profiles in families, Clin Genet 85 (2014) 433-440. [181] A. von Eckardstein, H. Holz, M. Sandkamp, W. Weng, H. Funke, G. Assmann, Apolipoprotein C- III(Lys58----Glu). Identification of an apolipoprotein C-III variant in a family with hyperalphalipoproteinemia, J Clin Invest 87 (1991) 1724-1731. [182] H. Liu, C. Labeur, C.F. Xu, R. Ferrell, L. Lins, R. Brasseur, M. Rosseneu, K.M. Weiss, S.E. Humphries, P.J. Talmud, Characterization of the lipid-binding properties and lipoprotein lipase inhibition of a novel apolipoprotein C-III variant Ala23Thr, J Lipid Res 41 (2000) 1760- 1771. [183] J. Crosby, G.M. Peloso, P.L. Auer, D.R. Crosslin, N.O. Stitziel, L.A. Lange, Y. Lu, Z.Z. Tang, H. Zhang, G. Hindy, N. Masca, K. Stirrups, S. Kanoni, R. Do, G. Jun, Y. Hu, H.M. Kang, C. Xue, A. Goel, M. Farrall, S. Duga, P.A. Merlini, R. Asselta, D. Girelli, O. Olivieri, N. Martinelli, W. Yin, D.

51 Reilly, E. Speliotes, C.S. Fox, K. Hveem, O.L. Holmen, M. Nikpay, D.N. Farlow, T.L. Assimes, N. Franceschini, J. Robinson, K.E. North, L.W. Martin, M. DePristo, N. Gupta, S.A. Escher, J.H. Jansson, N. Van Zuydam, C.N. Palmer, N. Wareham, W. Koch, T. Meitinger, A. Peters, W. Lieb, R. Erbel, I.R. Konig, J. Kruppa, F. Degenhardt, O. Gottesman, E.P. Bottinger, C.J. O'Donnell, B.M. Psaty, C.M. Ballantyne, G. Abecasis, J.M. Ordovas, O. Melander, H. Watkins, M. Orho- Melander, D. Ardissino, R.J. Loos, R. McPherson, C.J. Willer, J. Erdmann, A.S. Hall, N.J. Samani, P. Deloukas, H. Schunkert, J.G. Wilson, C. Kooperberg, S.S. Rich, R.P. Tracy, D.Y. Lin, D. Altshuler, S. Gabriel, D.A. Nickerson, G.P. Jarvik, L.A. Cupples, A.P. Reiner, E. Boerwinkle, S. Kathiresan, Loss-of-function mutations in APOC3, triglycerides, and coronary disease, N Engl J Med 371 (2014) 22-31. [184] A.B. Jorgensen, R. Frikke-Schmidt, B.G. Nordestgaard, A. Tybjaerg-Hansen, Loss-of-function mutations in APOC3 and risk of ischemic vascular disease, N Engl J Med 371 (2014) 32-41. [185] W.V. Brown, M.L. Baginsky, Inhibition of lipoprotein lipase by an apoprotein of human very low density lipoprotein, Biochem Biophys Res Commun 46 (1972) 375-382. [186] R.J. Havel, V.G. Shore, B. Shore, D.M. Bier, Role of specific glycopeptides of human serum lipoproteins in the activation of lipoprotein lipase, Circ Res 27 (1970) 595-600. [187] C.S. Wang, W.J. McConathy, H.U. Kloer, P. Alaupovic, Modulation of lipoprotein lipase activity by apolipoproteins. Effect of apolipoprotein C-III, J Clin Invest 75 (1985) 384-390. [188] A.G. Holleboom, H. Karlsson, R.S. Lin, T.M. Beres, J.A. Sierts, D.S. Herman, E.S. Stroes, J.M. Aerts, J.J. Kastelein, M.M. Motazacker, G.M. Dallinga-Thie, J.H. Levels, A.H. Zwinderman, J.G. Seidman, C.E. Seidman, S. Ljunggren, D.J. Lefeber, E. Morava, R.A. Wevers, T.A. Fritz, L.A. Tabak, M. Lindahl, G.K. Hovingh, J.A. Kuivenhoven, Heterozygosity for a loss-of-function mutation in GALNT2 improves plasma triglyceride clearance in man, Cell Metab 14 (2011) 811-818. [189] C.S. Gangabadage, J. Zdunek, M. Tessari, S. Nilsson, G. Olivecrona, S.S. Wijmenga, Structure and dynamics of human apolipoprotein CIII, J Biol Chem 283 (2008) 17416-17427. [190] J.T. Sparrow, H.J. Pownall, F.J. Hsu, L.D. Blumenthal, A.R. Culwell, A.M. Gotto, Lipid binding by fragments of apolipoprotein C-III-1 obtained by thrombin cleavage, Biochemistry 16 (1977) 5427-5431. [191] D.A. Lambert, A.L. Catapano, L.C. Smith, J.T. Sparrow, A.M. Gotto, Jr., Effect of the apolipoprotein C-II/C-III1 ratio on the capacity of purified milk lipoprotein lipase to hydrolyse triglycerides in monolayer vesicles, Atherosclerosis 127 (1996) 205-212. [192] M. Sundaram, S. Zhong, M. Bou Khalil, P.H. Links, Y. Zhao, J. Iqbal, M.M. Hussain, R.J. Parks, Y. Wang, Z. Yao, Expression of apolipoprotein C-III in McA-RH7777 cells enhances VLDL assembly and secretion under lipid-rich conditions, J Lipid Res 51 (2010) 150-161. [193] W. Qin, M. Sundaram, Y. Wang, H. Zhou, S. Zhong, C.C. Chang, S. Manhas, E.F. Yao, R.J. Parks, P.J. McFie, S.J. Stone, Z.G. Jiang, C. Wang, D. Figeys, W. Jia, Z. Yao, Missense mutation in APOC3 within the C-terminal lipid binding domain of human ApoC-III results in impaired assembly and secretion of triacylglycerol-rich very low density lipoproteins: evidence that ApoC-III plays a major role in the formation of lipid precursors within the microsomal lumen, J Biol Chem 286 (2011) 27769-27780. [194] E.D. Breyer, N.A. Le, X. Li, D. Martinson, W.V. Brown, Apolipoprotein C-III displacement of apolipoprotein E from VLDL: effect of particle size, J Lipid Res 40 (1999) 1875-1882. [195] N. Maeda, H. Li, D. Lee, P. Oliver, S.H. Quarfordt, J. Osada, Targeted disruption of the apolipoprotein C-III gene in mice results in hypotriglyceridemia and protection from postprandial hypertriglyceridemia, J Biol Chem 269 (1994) 23610-23616. [196] M.C. Jong, P.C. Rensen, V.E. Dahlmans, H. van der Boom, T.J. van Berkel, L.M. Havekes, Apolipoprotein C-III deficiency accelerates triglyceride hydrolysis by lipoprotein lipase in wild- type and apoE knockout mice, J Lipid Res 42 (2001) 1578-1585. [197] Y. Ito, N. Azrolan, A. O'Connell, A. Walsh, J.L. Breslow, Hypertriglyceridemia as a result of human apo CIII gene expression in transgenic mice, Science 249 (1990) 790-793.

52 [198] Y. Ding, Y. Wang, H. Zhu, J. Fan, L. Yu, G. Liu, E. Liu, Hypertriglyceridemia and delayed clearance of fat load in transgenic rabbits expressing human apolipoprotein CIII, Transgenic Res 20 (2011) 867-875. [199] J. Wei, H. Ouyang, Y. Wang, D. Pang, N.X. Cong, T. Wang, B. Leng, D. Li, X. Li, R. Wu, Y. Ding, F. Gao, Y. Deng, B. Liu, Z. Li, L. Lai, H. Feng, G. Liu, X. Deng, Characterization of a hypertriglyceridemic transgenic miniature pig model expressing human apolipoprotein CIII, FEBS J 279 (2012) 91-99. [200] T. Ebara, R. Ramakrishnan, G. Steiner, N.S. Shachter, Chylomicronemia due to apolipoprotein CIII overexpression in apolipoprotein E-null mice. Apolipoprotein CIII-induced hypertriglyceridemia is not mediated by effects on apolipoprotein E, J Clin Invest 99 (1997) 2672-2681. [201] H.V. de Silva, S.J. Lauer, J. Wang, W.S. Simonet, K.H. Weisgraber, R.W. Mahley, J.M. Taylor, Overexpression of human apolipoprotein C-III in transgenic mice results in an accumulation of apolipoprotein B48 remnants that is corrected by excess apolipoprotein E, J Biol Chem 269 (1994) 2324-2335. [202] R.L. Jackson, S. Tajima, T. Yamamura, S. Yokoyama, A. Yamamoto, Comparison of apolipoprotein C-II-deficient triacylglycerol-rich lipoproteins and trioleoylglycerol/phosphatidylcholine-stabilized particles as substrates for lipoprotein lipase, Biochim Biophys Acta 875 (1986) 211-219. [203] E.J. Schaefer, R.E. Gregg, G. Ghiselli, T.M. Forte, J.M. Ordovas, L.A. Zech, H.B. Brewer, Familial Apolipoprotein-E Deficiency, Journal of Clinical Investigation 78 (1986) 1206-1219. [204] R.W. Mahley, S.C. Rall, Jr., Apolipoprotein E: far more than a lipid transport protein, Annu Rev Genomics Hum Genet 1 (2000) 507-537. [205] T. Mazzone, C. Reardon, Expression of heterologous human apolipoprotein E by J774 macrophages enhances cholesterol efflux to HDL3, J Lipid Res 35 (1994) 1345-1353. [206] P.G. Yancey, H. Yu, M.F. Linton, S. Fazio, A pathway-dependent on apoE, ApoAI, and ABCA1 determines formation of buoyant high-density lipoprotein by macrophage foam cells, Arterioscler Thromb Vasc Biol 27 (2007) 1123-1131. [207] S.H. Zhang, R.L. Reddick, J.A. Piedrahita, N. Maeda, Spontaneous hypercholesterolemia and arterial lesions in mice lacking apolipoprotein E, Science 258 (1992) 468-471. [208] S. Bellosta, R.W. Mahley, D.A. Sanan, J. Murata, D.L. Newland, J.M. Taylor, R.E. Pitas, Macrophage-specific expression of human apolipoprotein E reduces atherosclerosis in hypercholesterolemic apolipoprotein E-null mice, J Clin Invest 96 (1995) 2170-2179. [209] S. Fazio, V.R. Babaev, A.B. Murray, A.H. Hasty, K.J. Carter, L.A. Gleaves, J.B. Atkinson, M.F. Linton, Increased atherosclerosis in mice reconstituted with apolipoprotein E null macrophages, Proc Natl Acad Sci U S A 94 (1997) 4647-4652. [210] S. Eisenberg, E. Sehayek, T. Olivecrona, I. Vlodavsky, Lipoprotein lipase enhances binding of lipoproteins to heparan sulfate on cell surfaces and extracellular matrix, J Clin Invest 90 (1992) 2013-2021. [211] P.C. Rensen, T.J. van Berkel, Apolipoprotein E effectively inhibits lipoprotein lipase-mediated lipolysis of chylomicron-like triglyceride-rich lipid emulsions in vitro and in vivo, J Biol Chem 271 (1996) 14791-14799. [212] M.C. Jong, V.E. Dahlmans, M.H. Hofker, L.M. Havekes, Nascent very-low-density lipoprotein triacylglycerol hydrolysis by lipoprotein lipase is inhibited by apolipoprotein E in a dose- dependent manner, Biochem J 328 ( Pt 3) (1997) 745-750. [213] Y.D. Huang, X.Q. Liu, S.C. Rall, J.M. Taylor, A. von Eckardstein, G. Assmann, R.W. Mahley, Overexpression and accumulation of apolipoprotein E as a cause of hypertriglyceridemia, Journal of Biological Chemistry 273 (1998) 26388-26393. [214] S. Calandra, C. Priore Oliva, P. Tarugi, S. Bertolini, APOA5 and triglyceride metabolism, lesson from human APOA5 deficiency, Curr Opin Lipidol 17 (2006) 122-127.

53 [215] C. Priore Oliva, F. Carubbi, F.G. Schaap, S. Bertolini, S. Calandra, Hypertriglyceridaemia and low plasma HDL in a patient with apolipoprotein A-V deficiency due to a novel mutation in the APOA5 gene, J Intern Med 263 (2008) 450-458. [216] L.A. Pennacchio, M. Olivier, J.A. Hubacek, J.C. Cohen, D.R. Cox, J.C. Fruchart, R.M. Krauss, E.M. Rubin, An apolipoprotein influencing triglycerides in humans and mice revealed by comparative sequencing, Science 294 (2001) 169-173. [217] P.J. O'Brien, W.E. Alborn, J.H. Sloan, M. Ulmer, A. Boodhoo, M.D. Knierman, A.E. Schultze, R.J. Konrad, The novel apolipoprotein A5 is present in human serum, is associated with VLDL, HDL, and chylomicrons, and circulates at very low concentrations compared with other apolipoproteins, Clin Chem 51 (2005) 351-359. [218] D.S. Cho, S. Woo, S. Kim, C.D. Byrne, J.H. Kong, K.C. Sung, Estimation of plasma apolipoprotein B concentration using routinely measured lipid biochemical tests in apparently healthy Asian adults, Cardiovasc Diabetol 11 (2012) 55. [219] X. Gao, T.M. Forte, R.O. Ryan, Influence of apolipoprotein A-V on hepatocyte lipid droplet formation, Biochem Biophys Res Commun 427 (2012) 361-365. [220] R.B. Weinberg, V.R. Cook, J.A. Beckstead, D.D. Martin, J.W. Gallagher, G.S. Shelness, R.O. Ryan, Structure and interfacial properties of human apolipoprotein A-V, J Biol Chem 278 (2003) 34438-34444. [221] N. Baroukh, E. Bauge, J. Akiyama, J. Chang, V. Afzal, J.C. Fruchart, E.M. Rubin, J. Fruchart-Najib, L.A. Pennacchio, Analysis of apolipoprotein A5, c3, and plasma triglyceride concentrations in genetically engineered mice, Arterioscler Thromb Vasc Biol 24 (2004) 1297-1302. [222] S. Formisano, H.B. Brewer, Jr., J.C. Osborne, Jr., Effect of pressure and ionic strength on the self-association of Apo-A-I from the human high density lipoprotein complex, J Biol Chem 253 (1978) 354-359. [223] F. Ferrato, F. Carriere, L. Sarda, R. Verger, A critical reevaluation of the phenomenon of interfacial activation, Methods Enzymol 286 (1997) 327-347. [224] F.S. Steven, M.M. Griffin, T.P. Hulley, P. Brooman, Interaction of alpha, beta-diethyl stilbestrol 4,4'-bisphosphate with arginyl substrates resulting in apparent inhibition of trypsin and thrombin, Eur J Biochem 125 (1982) 305-309. [225] L. Wang, D. Atkinson, D.M. Small, Interfacial properties of an amphipathic alpha-helix consensus peptide of exchangeable apolipoproteins at air/water and oil/water interfaces, J Biol Chem 278 (2003) 37480-37491. [226] R.C. Kowal, J. Herz, K.H. Weisgraber, R.W. Mahley, M.S. Brown, J.L. Goldstein, Opposing effects of apolipoproteins E and C on lipoprotein binding to low density lipoprotein receptor-related protein, J Biol Chem 265 (1990) 10771-10779. [227] U. Beisiegel, W. Weber, G. Bengtsson-Olivecrona, Lipoprotein lipase enhances the binding of chylomicrons to low density lipoprotein receptor-related protein, Proc Natl Acad Sci U S A 88 (1991) 8342-8346. [228] P. Tornvall, G. Olivecrona, F. Karpe, A. Hamsten, T. Olivecrona, Lipoprotein lipase mass and activity in plasma and their increase after heparin are separate parameters with different relations to plasma lipoproteins, Arterioscler Thromb Vasc Biol 15 (1995) 1086-1093. [229] M. Merkel, B. Loeffler, M. Kluger, N. Fabig, G. Geppert, L.A. Pennacchio, A. Laatsch, J. Heeren, Apolipoprotein AV accelerates plasma hydrolysis of triglyceride-rich lipoproteins by interaction with proteoglycan-bound lipoprotein lipase, J Biol Chem 280 (2005) 21553-21560. [230] X. Shu, L. Nelbach, M.M. Weinstein, B.L. Burgess, J.A. Beckstead, S.G. Young, R.O. Ryan, T.M. Forte, Intravenous injection of apolipoprotein A-V reconstituted high-density lipoprotein decreases hypertriglyceridemia in apoav-/- mice and requires glycosylphosphatidylinositol- anchored high-density lipoprotein-binding protein 1, Arterioscler Thromb Vasc Biol 30 (2010) 2504-2509. [231] J.S. Millar, D.A. Cromley, M.G. McCoy, D.J. Rader, J.T. Billheimer, Determining hepatic triglyceride production in mice: comparison of poloxamer 407 with Triton WR-1339, J Lipid Res 46 (2005) 2023-2028.

54 [232] J.M. Ong, P.A. Kern, Effect of feeding and obesity on lipoprotein lipase activity, immunoreactive protein, and messenger RNA levels in human adipose tissue, J Clin Invest 84 (1989) 305-311. [233] G. Boden, Obesity, insulin resistance and free fatty acids, Curr Opin Endocrinol Diabetes Obes 18 (2011) 139-143. [234] N. Hosogai, A. Fukuhara, K. Oshima, Y. Miyata, S. Tanaka, K. Segawa, S. Furukawa, Y. Tochino, R. Komuro, M. Matsuda, I. Shimomura, Adipose tissue hypoxia in obesity and its impact on adipocytokine dysregulation, Diabetes 56 (2007) 901-911. [235] B. Wang, I.S. Wood, P. Trayhurn, Dysregulation of the expression and secretion of inflammation-related adipokines by hypoxia in human adipocytes, Pflugers Arch 455 (2007) 479-492. [236] P. Gonzalez-Muniesa, C. de Oliveira, F. Perez de Heredia, M.P. Thompson, P. Trayhurn, Fatty acids and hypoxia stimulate the expression and secretion of the adipokine ANGPTL4 (angiopoietin-like protein 4/ fasting-induced adipose factor) by human adipocytes, J Nutrigenet Nutrigenomics 4 (2011) 146-153. [237] N. Malo, J.A. Hanley, S. Cerquozzi, J. Pelletier, R. Nadon, Statistical practice in high-throughput screening data analysis, Nat Biotechnol 24 (2006) 167-175. [238] A. Lookene, L. Zhang, M. Hultin, G. Olivecrona, Rapid subunit exchange in dimeric lipoprotein lipase and properties of the inactive monomer, J Biol Chem 279 (2004) 49964-49972. [239] K. Shirai, R.L. Jackson, Lipoprotein lipase-catalyzed hydrolysis of p-nitrophenyl butyrate. Interfacial activation by phospholipid vesicles, J Biol Chem 257 (1982) 1253-1258. [240] E. Leroy, N. Bensel, J.L. Reymond, A low background high-throughput screening (HTS) fluorescence assay for lipases and esterases using acyloxymethylethers of umbelliferone, Bioorg Med Chem Lett 13 (2003) 2105-2108. [241] W.J. Geldenhuys, D. Aring, P. Sadana, A novel Lipoprotein lipase (LPL) agonist rescues the enzyme from inhibition by angiopoietin-like 4 (ANGPTL4), Bioorg Med Chem Lett 24 (2014) 2163-2167. [242] G. Shattat, R. Al-Qirim, Y. Al-Hiari, G.A. Sheikha, T. Al-Qirim, W. El-Huneidi, M. Shahwan, Synthesis and anti-hyperlipidemic evaluation of N(benzoylphenyl)-5-fluoro-1H-indole-2- carboxamide derivatives in Triton WR-1339-induced hyperlipidemic rats, Molecules 15 (2010) 5840-5849. [243] K. Tsutsumi, Y. Inoue, A. Shima, K. Iwasaki, M. Kawamura, T. Murase, The novel compound NO- 1886 increases lipoprotein lipase activity with resulting elevation of high density lipoprotein cholesterol, and long-term administration inhibits atherogenesis in the coronary arteries of rats with experimental atherosclerosis, J Clin Invest 92 (1993) 411-417. [244] W. Yin, K. Tsutsumi, Z. Yuan, B. Yang, Effects of the lipoprotein lipase activator NO-1886 as a suppressor agent of atherosclerosis in aorta of mild diabetic rabbits, Arzneimittelforschung 52 (2002) 610-614. [245] W. Yin, K. Tsutsumi, Lipoprotein lipase activator NO-1886, Cardiovasc Drug Rev 21 (2003) 133- 142. [246] C. Zhang, W. Yin, D. Liao, L. Huang, C. Tang, K. Tsutsumi, Z. Wang, Y. Liu, Q. Li, H. Hou, M. Cai, J. Xiao, NO-1886 upregulates ATP binding cassette transporter A1 and inhibits diet-induced atherosclerosis in Chinese Bama minipigs, J Lipid Res 47 (2006) 2055-2063. [247] C. Fievet, B. Staels, Liver X receptor modulators: effects on lipid metabolism and potential use in the treatment of atherosclerosis, Biochem Pharmacol 77 (2009) 1316-1327. [248] G. Bengtsson, T. Olivecrona, Binding of deoxycholate to lipoprotein lipase, Biochim Biophys Acta 575 (1979) 471-474. [249] M. Makishima, A.Y. Okamoto, J.J. Repa, H. Tu, R.M. Learned, A. Luk, M.V. Hull, K.D. Lustig, D.J. Mangelsdorf, B. Shan, Identification of a nuclear receptor for bile acids, Science 284 (1999) 1362-1365. [250] D.J. Parks, S.G. Blanchard, R.K. Bledsoe, G. Chandra, T.G. Consler, S.A. Kliewer, J.B. Stimmel, T.M. Willson, A.M. Zavacki, D.D. Moore, J.M. Lehmann, Bile acids: natural ligands for an orphan nuclear receptor, Science 284 (1999) 1365-1368.

55 [251] I.J. A, N.S. Tan, L. Gelman, S. Kersten, J. Seydoux, J. Xu, D. Metzger, L. Canaple, P. Chambon, W. Wahli, B. Desvergne, In vivo activation of PPAR target genes by RXR homodimers, EMBO J 23 (2004) 2083-2091.